Orientation by jumping spiders of the genus Phidippus (Araneae, Salticidae) during the pursuit of prey


Material Information

Orientation by jumping spiders of the genus Phidippus (Araneae, Salticidae) during the pursuit of prey
Physical Description:
202 leaves : ill. ; 28 cm.
Hill, David Edwin, 1948-
Publication Date:


Subjects / Keywords:
Phidippus   ( lcsh )
Animal orientation   ( lcsh )
Zoology thesis Ph. D
Dissertations, Academic -- Zoology -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Bibliography: leaves 194-201.
General Note:
General Note:
Statement of Responsibility:
by David Edwin Hill.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 022951099
oclc - 05349225
System ID:

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
        Page iv
        Page v
    Chapter 1. Introduction
        Page 1
        Page 2
    Chapter 2. Materials and methods
        Page 3
        Page 4
        Page 5
        Page 6
    Chapter 3. Salticid vision
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
    Chapter 4. The economics of predation: Appetitive behavior
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
    Chapter 5. The predatory behavior of salticid spiders
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
    Chapter 6. Orientation turns directed toward the prey
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Chapter 7. Criteria for pursuit
        Page 47
        Page 48
        Page 49
    Chapter 8. Selection of the access route
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
    Chapter 9. Orientation with reference to the direction of the immediate route
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
    Chapter 10. Orientation with reference to gravity
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
    Chapter 11. Visual reference to plant configurations
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
    Chapter 12. Visual determinants of the reorientation turn
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
    Chapter 13. Orientation in three dimensions
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
    Chapter 14. Concluding discussion
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
    References cited
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
    Biographical sketch
        Page 202
        Page 203
        Page 204
Full Text








I would like to express my great appreciation for the insightful

work of Karl Hubert Heil of Frankfurt am Main, who probably perished

during the conflagration of World War II in Germany, long before I was

born. I have also been inspired by the careful and, indeed, ingenious

approach of Dr. Michael F. Land, currently of the University of Sussex,

to the study of salticid behavior. In addition I wish to thank the

members of my committee at the University of Florida, Dr. Jonathan

Reiskind (chairman), Dr. John F. Anderson, and Dr. James E. Lloyd, for

the encouragement and thoughtful criticism which they have provided.

I must also cite the members of my own evolutionary unit, including my

dear wife Rose Marie, and our daughters Jennifer, Sara, and Vanessa,

for their respective contributions. Last, but certainly not least in

any way, I acknowledge the consistent quality of the cooperation

provided by those individual Phidippus which receive mention in these




ABSTRACT . . . . . . .






INTRODUCTION . . . . . . . . . .

MATERIALS AND METHODS... .. . . .... ....

SALTICID VISION . . . . . . . .







(b ORIENTATION) . . . . . . .


(RADIAL ORIENTATION) . . . . . . .



REFERENCES CITED . . . . . . . . . .

BIOGRAPHICAL SKETCH . . . . . . . . .



. . . 64

. . 110

. . . 147

. . . 155

S. . . 176

. . . 189

. . . 194

. . . 202


Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy



David Edwin Hill

June 1978

Chairman: Jonathan Reiskind
Major Department: Zoology

Jumping spiders of the genus Phidippus tend to occupy waiting

positions on plants. From these reconnaissance positions, the spiders

are often required to utilize indirect routes of access (detours) to

attain a position from which sighted prey (primary objective) can be


Selection of an appropriate route of access is based upon movement

toward a visually determined secondary objective (plant configuration),

which may provide access to the prey position. In moving toward a

secondary objective, the spider must turn away from the prey. At all

times during a pursuit, the spider retains a memory of the relative

position of the prey in space, with respect to several spider-centered

reference systems which are available for the determination of

direction. This memory of prey position is frequently expressed in the

form of a reorientation turn directed toward the expected position of

the prey. Each reorientation turn initiates a new segment of the


Phidippus can employ the immediate direction (route) of pursuit,

the force of gravity, and a radial direction with respect to the route

perceived through visual cues (including those cues furnished by both

the background and the immediate plant configuration) as reference

directions, for memory of the relative position of the prey. Phidippus

also compensate for their own movement in these reference systems in

determining a reorientation direction.

The indirect pursuit of sighted positions is a general feature of

the movement of Phidippus in vegetation, and the same reference systems

for reorientation that are employed during the pursuit of prey can

assuredly be applied to visually directed movement in general.


Salticid spiders utilize their highly developed vision to locate,

evaluate, and pursue potential prey. Since most of these spiders dwell

on plants, where the approach to prey is seldom in the form of a

straight line, effective access to prey requires an ability to utilize

an indirect route of pursuit.

Heil (1936) was the first to appreciate the significance of such

detoured routes, in his studies of the salticids Evarcha marcgravi and

E. blancardi [currently E. arcuata (Clerck) and E. falcata (Clerck),

respectively]. Heil's insight into the remarkable ability of salticids

to locate and utilize a detour during the pursuit of prey probably owed

as much to his extensive observations of Evarcha in the field as it did

to his documentation of detour behavior in the laboratory. Unfortu-

nately, his observations of what he termed one of the higher accom-

plishments of the salticids have received no attention in the subsequent

literature. Apart from a singular account of the circuitous approach

of a Phidippus audax in pursuit of a grasshopper (Bilsing, 1920), there

is essentially no additional literature on the subject.

A salticid requires an accurate memory for the relative position

of prey in space if it is to utilize a detour of any complexity beyond

that involved in a simple approach where prey is subject to visual

survey at all times. The objectives of this study include the evalu-

ation of the ability of certain salticid spiders (Phidippus) to


determine and successfully negotiate detours during the pursuit of prey,

and the assessment of the forms of sensory information which contribute

to the orientation of these spiders during such a detour.


The species of Phidippus which have been observed during the

course of this study are listed in Table 1. All experiments were

performed with individuals of P. pulcherrimus, and many of these were

repeated with other species of Phidippus to extend the generality of

the results. Selection of P. pulcherrimus as the primary subject of

study was based upon the ease with which these spiders adapted to

laboratory situations, as well as their willingness to run pursuits

under artificial conditions. This is without doubt related to the

habitat of these spiders, which generally dwell upon herbacious plants

in an open field. By contrast, individual P. pulcher (Walckenaer)

appeared to be preoccupied with the problem of ascent and often refused

to pursue prey in laboratory situations. P. pulcher are generally

found living in trees.

Initially several sequences of pursuit behavior were recorded

during field observation of P. princeps. Subsequently, experimental

problems of prey pursuit have been utilized to increase the volume of

data available for statistical evaluation.

The basic technique employed in the laboratory in the study of

pursuit behavior and orientation is quite simple. Individual spiders

are placed upon structures of varying design, stimulated with a dead

fly cemented to the end of a long strand of hair (standard lure), and

certain aspects of the ensuing behavior are recorded in a manner which

Table 1. Species of Phidippus observed during the course of this study.



P. pulcherrimus
Keyserling, 1884

P. princeps
(Peckham, 1883)

P. regius
C. L. Koch, ]846

P. clarus
Keyserling, 1884

P. audax
(Hentz, 1845)

Immatures collected from
resting sacs on the herb
Euthamia minor at Big
Prairie, Ocala National
Forest, Marion County,
Florida. Reared to

Offspring of animals
captured at Big Prairie.

Immatures observed living
on Euphorbia esula herbs
at Groveland Field,
Hennepin County, Minnesota.

Collected at Groveland,
reared to maturity.

living on
plants at
reared to

as immatures
Big Prairie,

Immatures collected from
sweep of grasses and
herbacious plants at Big
Prairie, reared to maturity.

Adults observed along the
margin of Newnan's Lake, on
grasses and herbacious
plants. Alachua County,

females: 1-6
mal e: 7

females: 8-14
males: 15-18

female: 19
male: 20

females: 21-24
male: 25
(SN 21 and 25 are

males: 26-28

With each presentation of experimental results, individual spiders are
referred to by the numbers given at the right, above. Thus SN 24fREG
refers to spider number (SN) 24, a female P. regius. The first three
letters of each species name are included in the abbreviated title.


varies with the design of each experiment. If necessary, spiders are

lured into starting positions with the same stimulus. Many of the

experiments involve a determination of the direction faced by the

spider, by projection of the axis of the spider onto a visible scale.

This process is greatly facilitated by the relatively large size of

Phidippus jumping spiders, as well as by the presence of conspicuous

and symmetrical groups of light-colored scales upon the prosoma of each


Unless otherwise specified, artificial lighting was provided by

either 100 w or 150 w (about 1500 to 2500 lumens) incandescent light

bulbs, situated inside aluminum parabolic reflecting lamps placed 0.5

to 1.5 m above each experimental situation. Laboratory temperature

was consistently held near 250C.

As noted by Kastner (1950) in his study of Evarcha falcata,

individual salticids can have distinct and consistent "personalities,"

or individual modes of behavior as observed in a particular context.

Management of spiders for the purposes of this study has been on an

individual basis, as some individuals are markedly better performers in

a given experimental situation than are others. Similar variation in

the behavior of individuals was noted by Tryon (1940) in his study of

white rats. Hundreds of spiders have been screened in the selection of

the relatively few (N= 28) individuals which receive mention in this

study. This selection was based solely upon whether or not the spiders

would actively, and repeatedly, pursue a standard lure in laboratory

situations. Virtually all Phidippus observed in the course of this

study (including rejects) demonstrated an ability to run a detoured

pursuit, nonetheless.

After individual screening, some attention is devoted to the care

of "runners," to maintain their responsiveness at a healthy optimum. A

certain degree of food deprivation is required for this purpose, but

this cannot be carried to the extreme of starvation. The spiders are

kept, as adults, individually in Petri dishes. They are provided with

droplets of drinking water several times a week. The spiders which are

being used for an experiment are provided with flies (either Musca

domestic or vestigial-winged Drosophila melanogaster) and they are fed

only when the opisthosoma is markedly reduced in size and starvation

appears to be imminent.

Ordinary statistical methods, consistent with definitions provided

by Siegel (1956), Runyon and Haber (1971), and Mendenhall and Scheaffer

(1973), have been employed for the analysis of data. The chi test was

used for the analysis of simple enumerative data, as well as that

arranged in contingency tables. For presumed parameters, means are

generally presented + 1 standard error. Student's t test was utilized

to determine either the significance of the difference between two mean

values, or the significance of the difference between one mean and a

fixed value. Paired data were analyzed by means of linear least-

squares regression, the Pearson r correlation coefficient, and the z r

test for the significance of r.


As shown in Figure 1, salticids have eight eyes, six of which are

relatively large, situated upon the prosoma. Apart from the posterior

medial eyes (PME), for which a definite function has never been estab-

lished, there are basically two types of eyes: the anterior medial, or

principal eyes (AME), and the lateral eyes (ALE and PLE). The use of

separate eyes for different functions allows for a combination of high

resolution in a narrow field of vision provided by the AME (the retinae

of which function as the foveae of the spider) with a wide (3600) field

of peripheral vision and lesser resolution provided by the lateral

eyes (Homann, 1928; Land, 1969a, 1974).

The lateral eyes continuously survey a large field around the

spider; these function as movement detectors, capable of directing a

turn by the spider to face a moving stimulus with the AME (Petrunke-

vitch, 1907; Homann, 1928; Land, 1971; Eakin and Brandenburger, 1971;

Dill, 1975). These eyes have a single spectral class of receptors,

with a peak sensitivity at >= 536 nm in Menemerus confusus (Yamashita

and Takeda, 1976). In the PLE and part of the ALE, receptor separation

limits the angular resolution to about 1.00 of arc (Land, 1969a);

stimulation (by a decrease in light intensity) of only two receptors,

separated by at least 1.0 in succession can lead to a turn (facing)

toward the second of the two stimuli (Evarcha arcuata; Duelli, 1975).

The lateral eyes should have some additional, generalized functions,

Figure 1. Diagram of the approximate visual field of Phidippus as
viewed from above, showing the eyes and their respective optic nerves.
This diagram is drawn from information provided by Land(1969a), Duelli
(1975), and Hill (1975). The posterior lateral (PLE) and anterior
lateral eyes (ALE) provide the spider with a 3600 survey of the sur-
roundings; the region of binocular overlap of the ALE in front of the
spider is shaded. The diminuitive posterior medial eyes (PME) synapse
primarily with a group of interneurons of large diameter (Hill, 1975),
which may release a rapid escape response. The principal or anterior
medial eyes (AME) possess high resolution but a restricted visual
field, as shown in black. The tubular portion of each AME, including
the retina, can be moved from side to side to scan the area in front of
the spider as indicated by the arrows at lower right.

as each eye is connected equally to two completely different tracts of

interneurons (Hill, 1975).

The remarkably large AME of salticids are demonstrably used to

evaluate form in the determination of subsequent behavior, particularly

in the recognition of a potential mate or potential prey (Homann, 1928;

Heil, 1936; Kastner, 1950; Dzimirski, 1959). The bipolar sensory cells

of each AME are situated in 4 layers at the end of a long tube; the AME

bear fixed lenses but the eye tubes and retinae can be moved from side

to side and rotated to either scan or center upon stationary objects,

or to track moving objects (Land, 1969b). The field of vision is small

(Figure 1), but the resolution of these eyes is limited only by the

0.150 angular separation of the receptors, far less than the theoreti-

cal limit (0.50) for resolution by the compound eyes of insects (Land,

1969a, 1974). As with the lateral eyes, the depth of field of the AME

in Phidippus extends from a position nearer than 5 cm to the spider to

infinity (Land, 1969a).

Stratification of the retina of the salticid AME may relate to

color vision (Land, 1969a). Electroretinograms and direct measurement

of receptor potentials have established the presence of at least 3

distinct classes of receptors in the AME of P. regius (peak sensitivi-

ties at = 370 nm or UV, 532 nm or green, and 370-525 nm; DeVoe and

Zvargulis, 1967; DeVoe, 1975), and at least 4 spectral classes of

receptors in Menemerus confusus (peaks at = 360 nm or UV, 480-500 nm,

520-540 nm,and 580 nm; Yamashita and Takeda, 1976). Color recognition

should, then, play a considerable role in the evaluation of conspecifics,

potential prey, and even the immediate surroundings by the jumping

spider. Crane (1949) and Kgstner (1950) provide limited demonstrations

of the role of color vision in this regard..

The problem of distance resolution or depth perception is of

considerable importance with respect to the orientation of these

animals; distance information is not only a critical determinant of

many behaviors, such as the predatory jump, but it is also required

for the accurate appraisal of the relative prey position during a

pursuit run (see below).

Clearly distance resolution is a function of the front (face)

eyes, the AME flanked the the ALE. Of these, the ALE have the greatest

separation, as well as an overlapping field of vision (Figure 1),

suggestive of binocular vision. This role is supported by the work of

Homann (1928), who found that blinding of a single ALE reduced the

accuracy of a jump to the same extent as the blinding of both ALE. The

AME can also be demonstrated to provide a certain amount of distance

information (Homann, 1928; Heil, 1936). The greater resolution of the

AME is offset by the proximity of these eyes, however. Since points

more than 3 cm from the spider come to focus in a single receptor

layer of the AME, it is unlikely that a single AME can provide

monocular (focal plane) depth perception (Land, 1972b). Presumably,

in the intact spider, data from the fixed ALE and the moving AME are

superimposed (integrated) and comprise a consistent view of the visual

field to the front of the animal.

Compared with the wide-angle lenses of the PLE, the lenses of the

ALE are of significantly longer focal length, corresponding to a more

restricted field of vision to the front of the spider (Land, 1969a).

Peripheral angular resolution of the ALE (determined by separation of

the receptors) is comparable to that of the PLE (about 1.0 of arc),

but this increases to about 0.5 for the ALE receptors which receive

stimuli from the field of binocular overlap of the ALE, directly in

front of the spider (Land, 1972b, 1974). Since angular resolution is

critical to the resolution of distance, this feature agrees with the

role of the ALE in the binocular resolution of distance.

Figure 2 provides a simple analysis of the theoretical limit to

binocular resolution as a function of the separation of the eyes and

the angular resolution of each eye (ALE), for a point stimulus. The

length of each sampling interval increases greatly with distance

(corresponding to a decrease in the accuracy of distance information),

until an upper limit is reached at which no further distance can be

discerned from infinity. As noted in Figure 2, one cannot set precise

values for these sampling intervals, and they may vary greatly as

different combinations of one receptor stimulus from each eye are

compared within the central nervous system of the spider. Each

combination of this sort may be termed a distance sample, which by

itself can be expected to decline in accuracy as the distance from the

spider increases.

An object which subtends more than a single receptor from each eye,

then, can be expected to provide a series of distance samples of

varying accuracy. Statistical treatment of these data within the

nervous system of the spider could extend the accuracy of the distance

resolution considerably. The form of a stationary housefly (Musca

domestic) might subtend 15-20 receptors in each of the ALE at a

distance of 20 cm; without any sophistication for edge information this

could still provide the spider with 400 eye-to-eye combinations, or

distance samples, of varying accuracy. It follows that one cannot

develop a theoretical limit for the resolution of distance based

(a= 90, D= ) 4.6 5.7 7.6 11.5 (22.9) cm
L I I I i I I I I I
(a= 89.70, D= 40 cm) 5.0 6.4 8.9 14.6

Figure 2. Resolution of distance with binocular vision. This diagram
shows the retinae (R) and nodal points (n) of the ALE (large circles
at left). Each of the quadrangles (sampling units) in the visual
field to the right corresponds to the simultaneous stimulation of a
unique pair of receptors (one from each eye) as indicated by shading.
Given (J), the angular resolution of the eyes, and (d), the
separation of nodal points, one can determine a set of values for the
limits of the distance sampling intervals (D) as indicated in the
scale at the top of the diagram, by substituting integral values of
(b) into the equation given above. Each value of (a) corresponds to
an upper limit of (D); (a) should be chosen so that (90 -e) < (a)
(90o ). At the bottom of the figure are shown two possible scales of
boundary values for distance resolution by the ALE of Phidippus,
based upon W)= 0.5 and (d)= 0.2 cm, with (a) and the respective
upper limit of (D) indicated in each case in parentheses.

solely upon a point source of information and reasonably expect the

same limit to apply to a situation where many distance samples can be

made either simultaneously, or in succession. Prey moving across a

visual field could provide a large amount of potential data in a short

period of time. The configuration of plants near the prey position

could provide an enormous amount of distance information.

Many insects can utilize parallax gained from their own movement

to extend the range of their distance resolution (Horridge, 1977).

There is no reason to suspect that the same does not apply to a

salticid as it runs after prey. Salticids are not known to utilize

distinct movements in place, such as rocking from side to side, to

facilitate the resolution of distance. Phidippus generally do not move

the prosoma as they survey a potential prey for an extended period of

time. On occasion a spider will rotate about its axis, or tilt the

prosoma during a period of sustained visual survey (Figure 3). This

behavior could produce more binocular distance samples in certain

situations, but the actual function of this prosomal rotation is not



B 0

Figure 3. Phidippus before (A) and after (B) a prosomal rotation of
magnitude p When surveying an object, these spiders frequently tilt
the prosoma to the side, as if attempting to secure some additional
information. This behavior may be related to the evaluation of form
by the scanning AME, but it could also be employed to improve the
resolution of distance. Linear forms subtend more pairs of receptors
when they are aligned with the sagittal plane of the animal, for
increased distance sampling. Resolution of distance by Homo sapiens
is also related to the orientation of the object relative to the axis
of the head; the distance of a rod can be best determined by man when
the rod is vertical with respect to a vertical sagittal plane (Blake,
Camisa, and Antoinetti, 1976).


A predator may either search for prey in an active manner, or it

may occupy a waiting position until the prey moves into sight, and then

pursue the prey (Carthy, 1965). Although there is really a continuum

of potential behavior between these two extremes, it is clear that some

salticids move about actively throughout the day, while others,

including Phidippus, tend to occupy waiting positions (Enders, 1975).

Evarcha also assumes a waiting position in vegetation (Heil, ]936;

Kastner, 1950; Plett, 1962a). At the opposite extreme is Synageles, an

ant mimic, which moves about quickly with only brief intervals of pause

or visual survey (K'stner, 1950). Salticus (Kastner, 1950; Drees,

1952; Plett, 1962a, 1962b) and Sitticus (Kastner, 1950) are also

constant searchers.

Under certain conditions, however, it is possible for the waiting

Phidippus to become an active searcher. A female P. pulcherrimus

(SN IfPUL) ran after the standard fly lure in a series of pursuits,

returning after each pursuit to a waiting position, facing down in a

crouch near the top of the stem. During one pursuit, this individual

discovered an aphid among the terminal leaflets of a branch, and then

returned to the main stem to feed. After feeding, this spider repeated

the entire sequence three more times before it was interrupted, each

time moving out to the terminal leaflets of various branches of the

plant (Melilotus alba, or sweet clover) and conducting an extensive

tactile survey marked by a constant turning and tapping of the leaflets

with the forelegs, until an aphid was located and captured. It appears

that Phidippus can employ this alternate (searching) strategy with some

success when sedentary prey are available.

The adaptive significance of this versatility for the generalized

predator of insects is evident. When the available prey are moving

actively, or alighting on occasion, the spider can maximize its obser-

vation time, minimize its energy expenditure, and reduce the possibility

of being captured by other predators by remaining immobile in a

relatively concealed position, and pursuing prey as it appears in the

area of visual survey. This is precisely the behavior that typifies

many observations of Phidippus in the field. When sedentary prey are

available, however, waiting is of no avail for their capture; they must

be collected by a searcher.

P. princeps and P. pulcherrimus both tend to occupy waiting

positions near the top of herbacious plants in an open field. These

positions are generally on the main stem of the plant, facing down.

The waiting position must be of fundamental importance in determining

the ability of the spider to detect prey. Successful prey capture is

also a function of access, or the ability of the spider to reach the

prey before it escapes; access can also be related to the suitability

of a waiting position.

If prey tend to appear (alight) on the upper surfaces of leaves

near the top of a plant, then the spider can maximize its ability to

detect prey by assuming a position near the top of the plant, facing

down to survey the upper surfaces of the leaves. Regardless of where

prey tended to alight, they would be most visible from above (better

illumination). In addition, a spider that can jump a considerable

distance down to its prey has a greater chance of capturing that prey

if pursuit is initiated from a superior position. Thus access is also

facilitated by assuming a waiting position above the area of visual


If the spider waits at some distance from the main stem of a plant,

it is increasing the access time required to reach every other portion

of that plant by the amount of time required to reach the main stem

from its waiting position. Since access time is without doubt a

critical determinant of prey capture, this is a definite disadvantage.

The increase in access time results from the simple fact that the main

stem is the most direct avenue of approach, and in fact the only avenue

of approach, which physically connects the various portions of the

plant. As long as prey detection from a position on the main stem is

adequate, the spider should wait on the main stem.

By the foregoing analysis one could conclude that the spider can

maximize the number of prey captured per unit time by selecting a

waiting position near the top of a plant stem in an open field, facing

down; both prey detection and prey access should be optimal from this

position. Individual P. pulcherrimus which have been reared for their

entire lives within Petri dishes will assume and pursue prey from such.

a position if placed on either a vertical rod or a suitable plant.

Gardner (1965) observed the use of a similar waiting position by

P. coccineus. Greenquist and Rovner (1976) also observed a tendency

for certain lycosid spiders to orient in a vertical direction, facing

down, on artificial foliage. An additional feature of this position

lies in the fact that the spider is facing in the direction from which

other wandering spiders would necessarily approach to gain access to

positions above the waiting spider; essentially (and particularly when

the upper portion of the plant is isolated in an open field) the spider

stands guard at the gateway to the upper portion of the plant. This

may place the spider on a route utilized by potential prey, and it

should at least remove the danger that another salticid is going to

attack from above, since this area is presumably secure. If plants are

more crowded, it could at least reduce the danger of such an attack.

Prey detection (in terms of the number of prey sighted per unit

time) is also a function of the local prey density. If prey do not

appear after the spider assumes a particular waiting position, then it

is clearly of advantage for the spider to relocate. This response to a

lack of prey is undoubtedly widespread among the sedentary spiders; as

an example, Turnbull (1964) found that the theridiid Achaearanea

tepidariorum will search for a new web site if prey are lacking in a

given location. Plett (1962a) found that food deprivation led to

greater movement by Salticus scenicus.

Likewise, individual Phidippus princeps with large opisthosomata

(an index of the degree of satiation) tend to remain within or near

their resting sacs (near the top of plants) throughout the day, while

other individuals often move between waiting positions, far from any

visible resting sacs, many times during the course of a day (Hill,

1977b). This may be viewed as a functional response to exploit regions

of high prey density, since the net effect of this behavior is the

concentration of the spiders in those areas where prey is most


In summary it should be noted that Phidippus, like the other

salticids, is a diurnal hunter. Individuals of species which inhabit

the herbacious plants of an open field tend to emerge from their

resting sacs in the morning of suitable days to occupy a series of

waiting positions before returning to a resting sac (or constructing a

new one) for the night. The reliance of these spiders upon visual

survey from a waiting position for the detection of prey implies that

the ability of these spiders to pursue that prey along an available

route of access is of paramount importance to their success.


If an insect appears on a flat surface, near a salticid spider

waiting on that same surface, the spider will quickly turn to face the

insect with the AME. If an attack follows in this situation, the

spider will advance directly toward the prey at a variable but steady

pace, eventually slowing to a stalk or leg by leg advance in a

partially crouched position (legs drawn near to the body, thereby

reducing the apparent size of the spider as viewed by the prey), with

the pedipalps flickering up and down in unison. When a certain

distance from the prey is attained, the spider crouches and prepares to

jump by forming an attachment disk. In the case of Phidippus, the

spider then raises legs I and II toward the prey, spreads legs III

forward for balance and support during thrust, and then bends the

jumping legs IV against the substratum in preparation for the moment of

attack. A sudden extension of legs IV hurtles the spider into the air,

toward the prey, which is impaled between the fangs and grasped with

legs I and II.

Most of the descriptions of this simple attack sequence which are

available in the literature are based upon the observation of captive

salticids upon flat surfaces (Homann, 1928; Drees, 1952; Gardner, 1964,

1965; Hollis and Branson, 1964; Forster, 1977). Although the flat

surface may be relevant to the topography encountered by a Salticus

scenicus living on a wall during the pursuit of prey, it is not

representative of the plant configurations used for access to prey by

plant-dwelling Phidippus. Certain misconceptions, and particularly the

incorrect notion that salticids require continuous visual (AME) contact

with the prey during pursuit, can arise when one attempts to generalize

from observations of attack behavior conducted exclusively upon flat

surfaces. The direct pursuit which is described in this situation is

obviously the best solution to the problem of access, since direct

access is possible. Even in this situation, however, Phidippus may

side-step around the prey to attack crawling prey just behind the

moving front end. The usual, direct approach to relatively immobile

prey (such as a fly which has alighted) is shown in Figure 4 (A, D);

Also figured are typical sequences of pursuit which can be readily

demonstrated with simple modifications of the relative positions of

prey and routes of access to the position of a spider upon a flat


Although the problems of pursuit which are shown in Figure 4 are

rather simple, the basic features of pursuit which appear as a result

have not been described previously. When a direct approach to the

sighted prey (the primary objective) is not possible, the spider can

alternately pursue a visible route of access (secondary objective)

which is related in some way to the prey position. During this

indirect pursuit, or detour, the spider retains a memory of the rela-

tive position of the prey in space; the spider can integrate information

gathered during the initial observation of the prey with information

pertaining to the direction and extent of its own movement (movement

compensation) to determine the direction of the prey from a new posi-

tion. This determination is frequently expressed in the form of a


fA&I I


5 4 5
2 51 :'

Figure 4. Behavior of Phidippus during prey pursuit upon flat horizon-
tal (A-C) and vertical (D-F) surfaces. Even in relatively simple
situations, one may discern the basic features of pursuit which are
relevant to more complicated plant configurations. Each open circle
represents a spider position; the black spots represent the position of
prey, which consists of the standard dead fly dangling upon a long
hair. A, D: These simple demonstrations illustrate the direct
approach of the spider (vector of movement) toward prey resting upon
the same surface, or near the surface; this direct approach is
generally depicted as the typical or stereotyped response of a salticid
to prey. B, E: With the addition of a wooden rod projecting from the
surface, and the appearance of prey well away from that surface, the
basic features of indirect pursuit can be observed. These include the
initial turn toward the prey (1) to establish a- primary objective,
subsequent orientation toward a secondary objective 2) which comprises
part of the route of access, a rapid pursuit oriented (visually)
directly toward the base of the secondary objective (3; the base of the
secondary objective could be considered to be a tertiary objective, en
route to attainment of the secondary objective, en route to attainment
of the primary objective, the prey), followed by ascent (4) to the
secondary position and reorientation (5) to the primary objective. C:
On a horizontal surface, the spider turns to the prey (1), turns away
to run as direct a pursuit as is possible in this situation (2), and
then reorients (3) to the prey. F: Behavior on a vertical surface is
markedly different from that on a horizontal surface. After a turn to
prey (1), the spider ascends, relative to the prey (2), reorients (3),
side-steps (4), and then prepares to jump from above the prey (5).

reorientation turn toward the expected direction of the prey when the

spider has attained a new position. Evidence for this interpretation

of the indirect pursuit behavior of Phidippus will be presented in

subsequent sections, along with an analysis of the types of information

which can comprise a memory of the relative position of the prey by the

spider, during its pursuit of that prey.

At times a spider will approach prey indirectly while maintaining

visual (AME) contact with the prey, either by walking sideways (side-

stepping), or by walking in a manner intermediate between forward and

lateral walking. This behavior is most often observed when the spider

is close to its prey. To utilize a more rapid forward run during an

indirect pursuit, the spider is required to turn away from the prey,

since both legs and eyes are situated on the same body unit, the

prosoma. In addition, visual orientation to the indirect route of

access itself, which is most evident when a jump is required along that

route, precludes the use of the AME to continuously monitor the prey

position during pursuit.

Implicit in this discussion is the notion that detours provide the

only access to most of the area subject to visual survey by a waiting

Phidippus (or Evarcha; Heil, 1936). Even with a fairly direct route of

access, the actual prey position may be obscured by a leaf or stem

during part of the pursuit, forcing the spider to rely upon something

other than continuous visual (AME) contact to maintain its memory of

the prey position. Certain forms of concealed approach (see below)

which are utilized by Phidippus also require that these spiders turn

away from the prey during pursuit.

Representative sequences of the indirect pursuit of prey by

Phidippus are described in Figures 5-10. In each of these cases, the

reorientation turns are discrete events, each executed with a single

smooth movement which brings the spider to face its primary objective,

or at least a fairly accurate estimate of the actual position of that

objective. Figure 11 shows how a spider can attain a rather distant

objective by interrupting the pursuit with a series of reorientation

turns. Each reorientation turn initiates a new segment of the pursuit,

and a series of segments may be required in certain situations to bring

the spider into a position from which the prey can be captured.

Presumably these reorientation turns are used to correct and to main-

tain an accurate memory of the prey position during the pursuit; like

the initial orientation to prey, each reorientation can be considered

to affirm the relative position of the primary objective. It is

expected that there is a limitation to the ability of the individual

Phidippus to maintain a memory of prey position in the absence of

information provided by the object (perhaps a plant configuration

associated with the prey) of reorientation, particularly when a suc-

cession of detours within the context of a detoured pursuit are

required. Reorientation can also provide the spider with an opportu-

nity to assess the merit of continuing its pursuit.

Figure 12 shows the indirect pursuit of a potential mate by a male

P. audax. The use of detours by Phidippus is not restricted to the

problem of access to prey, but applies more generally to the movement

of these spiders in vegetation. Movement of both P. princeps and

Eris marginata (Hill, 1978) in the field is frequently interrupted by

periods of extensive turning, ostensibly for purposes of visual survey

5 cm

Figure 5. Indirect pursuit of prey by an immature P. princeps observed
in an old field habitat, based upon field sketch and measurements of
the configuration. Initially (1), the spider (open circles) rests in a
waiting position beneath a Solidago leaf. Then it turns (2) to face a
fly (F) which has just alighted at a distance of about 10 cm, upon a
branch of Euphorbia esula. Pedipalps flicker up and down in unison as
the spider surveys the fly. Subsequently the spider runs (3), reorients
in a single smooth turn to the position of the immobile fly (4), runs
(5), again reorients accurately (6), then climbs over a series of
Euphorbia leaves (7) to attain a position (8) facing the fly just as it
prepares to fly away. Presumably other pursuits were more successful
than this one. The young individual was about 5 mm in length. The
native P. princeps appear to thrive upon the introduced herb E. esula
in northeastern North America. After Hill (1977b).


5 cm

Figure 6. Indirect pursuit of an immobile (dead) fly by a female
P. pulcherrimus. Initial movement of the fly (black spot, as in subse-
quent figures) leads the spider to make its initial orientation to the
primary objective (1), followed by a steady, winding descent (2) to the
main stem of this sweet clover (Melilotus alba). After a brief stop,
the spider ascends steadily (3) on a winding course, stopping (4) to
turn toward the completely immobile fly. A slow, step by step stalk
with pedipalps flickering up and down in unison (5) is followed by a
short jump and capture of the fly. This particular pursuit is somewhat
unusual in its lack of intervening reorientation turns, but it does
show the extent to which a Phidippus will backtrack to attain an other-
wise inaccessible position. Since this fly was somewhat obscured by
leaflets, and completely immobile during the reorientation (4), it is
likely that this turn was initiated by the spider before it once again
sighted the fly.

sunlight, 240C

10 cm

Figure 7. Indirect pursuit of fly (standard lure) on sweet clover by a
female P. pulcherrimus. A standard procedure of removing the fly
stimulus as soon as the spider turned away in pursuit (2) was used here.
The spider orients to the fly (1), turns down (2), jumps (3), runs
quickly (4), turns toward the starting position (5), turns to a nearby
leaflet (6), jumps to that leaflet (7), runs quickly (8, 9), turns to
leaflet where fly appeared initially (10), runs (11), and again turns
to face the leaflets which may be associated with the sighted prey.
The "deliberate" appearance of the entire pursuit, and particularly the
turns toward the apparent objective (10, 12), suggest that the latter
were real examples of reorientation in the course of a sustained pursuit.
The use of a turn-down jump (2, 3) is frequently observed during an
indirect pursuit. The distinction between walking and running is based
solely upon a subjective impression of the velocity of the spider.


5 cm

Figure 8. Indirect pursuit of fly on Solidago by a female P. pulcher-
rimus as viewed from above. In this situation most of the movement of
the spider takes place in a single, horizontal plane; thus orientation
to gravity cannot account for the ability of the spider to relocate the
prey position. The dead fly is removed as the spider turns away from
its initial orientation (1). Then the spider walks steadily along a
route which includes several 90 turns (2), reorients to a leaf
apparently associated with the primary objective (3), turns down to an
accessible (nearer) leaf (4), jumps to that leaf (5), and reorients
immediately to the objective leaf (6). Notice that after reorientation
(3), a second segment of detour (4, 5) is utilized with no loss of
orientation by the spider.

5 cm

/SN 6fPuL

Figure 9. Indirect pursuit of sighted prey by a female P. pulcherrimus.
The spider turns to the fly (1), runs (2; the fly is removed at this
time), stops and attempts to reach a nearby leaflet with the forelegs,
with no success after about 5 sec (3), walks steadily (4), reorients
(5), walks (6), and then makes a series of small turns (7). The small
turns are apparently employed for visual (AME) survey of a restricted
area. Often a series of larger turns, encompassing 360, are observed
during a pause in the movement of a salticid spider. Notice that even
after the distraction of reaching behavior (3), the spider maintains a
memory of the prey position and continues its pursuit (4). Indirect
pursuit is not a stereotyped sequence of stimuli and simple responses
which can be readily disrupted by a subsequent distraction. It is
better viewed as the opportunistic exploitation of potential routes of
access by a spider which maintains an accurate, and persistent, memory
of the primary objective.


5 cm

Figure 10. Indirect pursuit of sighted prey by a female P. pulcher-
rimus. This sequence includes the initial turn to the fly (1), a fast
walk under the branch as the prey is removed (2), reorientation (3),
another fast walk under the branch (4), and three (5-7) intervals of
a "jerky walk" pattern of ascent. The significance of this staccato
stepping pattern is not known, but the visual effect is similar to
that of a hunting wasp on the prowl. The spider persists in its pur-
suit (4) after failing to gain access to the stem by venturing out to
position (3). In a similar pursuit, another individual made a series
of 4 forays out to the side branches until a position allowing for a
jump to the objective plant was attained.

15 N


10 cm


Figure 11. Indirect pursuit of sighted prey on an artificial plant
(constructed with wood dowels) by a particularly adept female
P. pulcherrimus. The fly was removed as soon as the spider turned away
from the initial orientation (1). The sequence was then: run (2),
reorient (3), turn down (4), jump (5), reorient (6), turn to green
paper "leaf" at the end of a near "branch" of the objective plant (7),
turn down (8), reorient (9), turn to near leaf (10), jump (11),
reorient (12), fast walk (13-15). Once attaining the objective position
this spider performed an extensive series of small turns. The sequence
includes 4 discrete reorientations to the objective leaf (1, 3, 9, and
12), each of which initiates a new segment of pursuit. Most individuals
tested did not complete this entire sequence, although they could
usually get as far as reorientation (9).


Figure 12. Indirect pursuit of a female P. audax by a male of the same
species, based upon a field sketch of the event. The sequence was:
(1) The male displays vigorously toward the female with legs I out-
stretched (note the symbolic representation above); the female faces
the male. The male continues to display (2) while side-stepping and
then (3) turns to jump to a grass stem. After reorientation to the
female (4), which has followed the movements of the male, the male
continues to display as he ascends the stem (5). After turning to run
quickly over the top of a broad blade of grass (6), the male returns to
the edge of the grass blade to look down from an appropriate position
(7). Upon sighting the display of the male above her, the female runs
in pursuit (8), and the male, in response, immediately turns around and
takes a long flying leap (free-fall, no drag-line is played out) away
from the pursuing female (9). The rapid escape of the male was quite
adaptive in this situation, as female P. audax which have mated will
never mate again, but instead they will pursue and devour the courting
males whenever possible. In these circumstances, the ability of the
male to retain a memory of the relative position of the female is
quite as important as his ability to assess her mood. Willing females
do not pursue, but wait. This fortuitous observation was made possible
by an unusually high density of adult P. audaxnear Newnan's Lake in
Alachua County, Florida during the fall of 1977.

of the surroundings. Often these spiders attain positions which were,

previously, the objects of visual survey, and presumably detours are

utilized for this purpose. In this study, the pursuit of prey is

separated from other, more general problems of directed movement only

because the presentation of prey to the spider allows one to manipulate

the objectives of the spider in a controlled fashion. Actually, even

within the context of a predatory pursuit, the spiders are concerned

with the problem of attaining sighted plant positions which may be

associated with the prey (Figures 7-11), just as they pursue secondary

objectives (also plant configurations) indirectly during the course of


In summary, the use of a detour can be viewed as an essential

feature of the directed movement of Phidippus upon vegetation, and this

behavior can be readily demonstrated when the spider is presented with

a specific problem of prey access. Use of a detour requires both an

ability to determine an appropriate route of access, and an ability to

retain a memory (compensated for movement by the spider) of the

relative position of the objective. This memory of prey position is

frequently expressed as a reorientation turn directed toward the

expected position of the prey.


Salticid spiders turn to face moving objects which appear within

the extensive visual field of the lateral eyes. A subsequent period of

visual survey of potential prey with the front eyes (AME and ALE),

which may occupy only a fraction of a second, necessarily precedes the

setting of the prey position as the primary objective, or the initia-

tion of the pursuit. As indicated in Figures 13-15, these directed

turns toward visual stimuli include both o< (stepping) and A (eleva-

tion of prosoma) components of movement.

Land (1971) demonstrated the ability of salticids to execute an

accurate o< turn to face a visual stimulus, in the absence of visual

feedback (open loop; Mittelstaedt, 1962, 1964); these spiders require

only the directional information provided by the initial sighting of a

stimulus by a lateral eye to determine the extent of this orientation

turn. In his demonstration, Land observed the stepping movement of

each 0 turn indirectly, by measuring the angle through which a spider,

affixed by the prosoma, would turn a paper ring held by the legs.

Figures 16 and 17 provide an additional demonstration of the ability of

untethered Phidippus to execute an accurate OC turn in the absence of

visual feedback from the stimulus during the course of the turn. In

this demonstration, the possibility that the spider might utilize tac-

tile information provided by the edge of a paper ring to monitor the




Figure 13. The OC turn. Adult male P. pulcherrimus before (A) and
after (B) completing a turn of magnitude OC in the horizontal plane.
The convention adopted for the sign (+ or -) of c< is shown in (A). A
stepping pattern like that described by Land (1972a) is usually
involved in an c, turn, although the spider can make small 06 turns in
place by pivoting.

Figure 14. The 6 turn. Phidippus before (A) and after (B) completion
of a rotation of the prosoma in the sagittal plane of magnitude 18.
Generally stepping is seldom utilized in the execution of a R turn,
which involves a pivot of the prosoma upon the legs in position, to
either raise (+ magnitude) or lower (-) the prosoma with respect to the
substratum. Phidippus can turn to look directly up from a surface in
this manner; by raising the pedicel above the substratum they can also
lower the front of the prosoma to look down from the edge of a surface.

Figure 15. Smooth execution of a turn (A to B) involving both oe and
A components of movement by an adult female P. pulcherrimus (traced
from photographs). The ability of Phidippus to execute a turn in three
dimensions, involving the coordinated activity of muscles concerned
with elevation of the prosoma (, ), as well as those concerned with
stepping upon an irregular surface (cV), can be readily demonstrated by
the presentation of a fly stimulus to the spider. Presentation of prey
above and behind a spider resting on a horizontal surface can elicit a
180 c>x turn coupled with an appropriate (+),B turn, as a single
smooth movement. Often a small 8 correction (presumably the result of
visual feedback from the prey) is required to face the prey directly,
at least when the spider is resting upon a flat surface. This is
related to the tendency of these spiders to turn a smaller R turn in
the absence of visual feedback than is required to face the prey.




Figure 16. Apparatus used to measure the ability of untethered
Phidippus to complete an accurate ox turn to face prey in the absence
of visual feedback, viewed from above (A), and in vertical section (B).
The entire apparatus is constructed of heavy white paper. The spider
(open circle) rests on a small (2 cm diameter) circular platform at the
center of a large cylinder. A numbered scale (F), used to measure the
direction faced by the spider, encircles the prey position. As shown
in (A), the direction initially faced by the spider (1), the position
at which prey appears (2), and the direction faced by the spider after
the orientation turn (3) are recorded in sequence by an observer
looking directly down from above the cylinder. As shown in (B), the
prey (a standard lure) is dropped into the trough well before the
spider has a chance to complete its turn. When there is any doubt in
this regard, the trial is not recorded. From (1), (2), and (3), values
of C> c, or the magnitude of O(required to face the original prey posi-
tion, and the actual value of C< elicited by the stimulus are deter-
mined as shown in (A).

06- C'


* r (for absolute
< 0,

-90 (oz test:

< 90- values)= 0.98
"90-^i~ .z test:
y P(r= 0) 0.0001


-180- 0 9b 08
Figure 17. Data collected from a single series of trials using the
technique described in Figure 16: Oe as a function of oe' This female
P. pulcherrimus has the ability to complete a remarkably Caccurate turn
toward the original position of a stimulus sighted by the lateral eyes
in the absence of visual feedback. Although comparable with similar
results obtained by Land (1971), none of the small turns in response to
a prey stimulus which he recorded have been observed. An interesting.
feature of this behavior lies in the fact that these spiders will always
make the shortest of two possible turns toward a stimulus; where a 2200
turn would suffice, the more direct -1400 turn is always utilized, as an
example. The accuracy of required determinations of the direction faced
by the spider from a scale 15 cm distant was frequently checked by com-
parison of an observed orientation (by the scale) with the actual posi-
tion of a fly which the spider was facing directly; the method is quite
reliable, + 50.

* *
S / 0


-180 -90 0 90 1800
Figure 17. Data collected from a single series of trials using the
technique described in Figure 16: O.^ as a function of 0< This female
P. pulcherrimus has the ability to complete a remarkably accurate turn
toward the original position of a stimulus sighted by the lateral eyes
in the absence of visual feedback. Although comparable with similar
results obtained by Land (1971), none of the small turns in response to
a prey stimulus which he recorded have been observed. An interesting
feature of this behavior lies in the fact that these spiders will always
make the shortest of two possible turns toward a stimulus; where a 2200
turn would suffice, the more direct -140 turn is always utilized, as an
example. The accuracy of required determinations of the direction faced
by the spider from a scale 15 cm distant was frequently checked by com-
parison of an observed orientation (by the scale) with the actual posi-
tion of a fly which the spider was facing directly; the method is quite
reliable, +_

extent of a turn (a potential factor in Land's demonstration, although

it does not detract from his demonstration that visual feedback is not

required) is not allowed. A similar demonstration of the accuracy of

A turns directed toward a stimulus in the absence of visual feedback

during each turn is shown in Figures 18-20; these turns consistently

fall short of the stimulus position. The execution of combined OC-A

turns in the absence of visual feedback can also be demonstrated, but

the accuracy of these turns (involving two components of direction)

has not been measured.

The fact that visual feedback is not required for an accurate turn

does not imply that visual feedback is not operative in non-experimen-

tal situations. Indeed, the very purpose of the orientation to prey

is to allow the spider to gather additional visual information with the

greater resolution and binocular vision afforded by the front eyes. If

visual feedback is available, it will be utilized by the spider to

direct subsequent behavior. When a second presentation of the prey

stimulus, in the initial position, is made after the completion of an

orientation turn executed in the absence of visual feedback by a spider

on the apparatus shown in Figure 16, the spider quickly makes a second

correcting turn toward the visible prey (unless the accuracy of the

initial turn was sufficient). Whenever the prey remains visible during

the orientation turn, the spider invariably turns directly toward the

prey, although the completion of this turn often requires a small but

distinct correcting turn; this correcting turn, which may be in a

direction opposite to that of the original turn, is clearly the product

of visual feedback. Even when running after moving prey, a Phidippus

can execute a continuous series of accurate orientation turns which are

Figure 18. Apparatus described in Figure 16, after modification for
the observation of 8 turns, as viewed from above. A 2.5 cm wide
vertical platform (at center) provides a waiting position for the
spider. Before each trial, the spider is shown a prey stimulus (the
standard lure) at (1) to return the spider to a standard starting
orientation. Following appearance of a stimulus at (2), the spider
turns (3). From observed values of (1), (2), and (3), both 8 c' or
the turn required to face the prey position, and the actual turn (B)
are determined. As in the observation of 0^, the fly was dropped so
that it could not be employed by the spider to monitor the extent of
the turn. Determination of the direction faced by the spider was
consistent with the vector shown in Figure 14, as well as with the
appearance of the spider when it was surveying the fly directly at
position (1), above. Nonetheless it is likely that measurement of
direction is less accurate here than in a situation where the spider is
viewed from above.

9001 1 ____

N= 103
r= 0.87
z test: P(r= O)z 0.0001
T-TI = -18.1 + 0.90
C ______
t test: P(,,8 c= 0)< 0. 000 1

I *
-J / I ^
/ "i/ I *'

I 30 0
I { .2 4 ..

0 60^ SN 5fPUL
0 30 60 90

Figure 19. Performance of a female P. pulcherrimus on the apparatus
shown in Figure 18. Although the angle turned by the spider correlates
significantly with the angle required to face the stimulus, the angle
turned is nonetheless significantly smaller than the required angle.
Many of the turns shown above consisted of 2 visibly distinct A compo-
nents executed in rapid succession. Subsequent presentation of prey in
the initial position could often elicit an additional, small A turn
(correction of the orientation to prey with visual feedback) to face
the prey. In a manner consistent with the format of subsequent figures
the dashed line represents the linear least-squares regression of Y (in
this case 8) on X (Ac).

900 ,I ii I II

N= 148 /
r= 0.90 /
z test: P(r= 0) 40.0001 / ,
-/A = -11.9 + 0.7 /0
t test: P(f{-i 0= 0)<0.0001 / /

60 / .

I." -"

9 -//1' I

30 :

H -

/i i '
07~ 3 0 0_________________

0 30 60 90


Figure 20. Performance of a male P. pulcherrimus on the apparatus
shown in Figure 18. The results are consistent with those shown in
Figure 19.


directed toward that prey (Figure 21). These may be closed-loop turns

directed by the AME (Land, 1972b).

Salticids also appear to turn frequently in the absence of

specific visual stimuli. When walking from plant to plant, both Eris

(Hill, 1978) and Phidippus will frequently stop to perform an extensive

(encompassing 3600) series of turns. More restricted bouts of visual

survey, consisting of a series of small turns directed toward the

vicinity of a particular plant configuration, are also observed; these

turns are presumably directed by the AME. Salticids readily turn to

face immobile objects, including prey, when these are in the field of

vision of the AME.

The partial turns which Land (1971) observed in response to a

directional stimulus probably correspond to these small survey turns.

These incomplete turns have not been observed in response to a specific

directional stimulus in any of the experimental situations utilized in

the course of this study. Extensive field observations of both Eris

and Phidippus suggest that there is no real basis for Land's attempt to

justify these partial turns in terms of energy conservation by the

spider; salticids turn frequently, and most of their turns are directed

toward inanimate objects, such as leaves blowing in the wind. Partial

turns in response to specific directional stimuli are not observed, as

a rule.

Many features of the orientation turn are shared with the reorien-

tation turn characteristic of the indirect (detoured) pursuit. In both

cases the turn is directed toward the expected position of prey as

accurately as possible, and both turns involve the simultaneous (or, in

some cases, sequential) completion of both c< and A components of

Figure 21. This diagram shows the approximate relationship of spider
(open circles) to prey (solid dots) positions during a running pursuit
of the moving prey. Phidippus can continuously orient to the moving
prey during its own movement. Stepping patterns involved in the move-
ment described above must include integrated components of both forward
movement (taxis) and turning (orientation). Observations of the
ability of Phidippus to capture moving prey suggest that these spiders
may continuously orient to a position in front of the moving prey in a
situation like this.


movement, generally integrated into a single, smoothly executed turn.

In addition, the accurate completion of both orientation and reorien-

tation turns requires an ability of the spider to account for the

relative starting positions of the components of its own body (proprio-

ception, or kinesthesis) in determining the appropriate motor output.

In the absence of visual feedback, proprioceptive feedback is presum-

ably required to monitor the extent of these directed turns.


Subsequent to the completion of an orientation turn toward a

particular stimulus, the salticid is in a position to evaluate the

object of its attention with the high resolution provided by the AME,

in addition to information provided by the ALE. The movement, size,

and form of potential prey all affect the probability of pursuit

(Plateau, 1887; Heil, 1936; Drees, 1952; Gardner, 1966; Land, 1971;

Dill, 1975). Features of either the behavior or appearance of prey can

be important determinants of the specific response of the spider as it

approaches that prey: At a distance of 1.5 cm, a stalking Evarcha will

subsequently avoid a bee, jump upon a fly, or subject a beetle to exten-

sive tactile exploration (Dahl, 1885). Salticus scenicus will stalk

after slowly moving prey, but they will run if the prey is moving

quickly (Drees, 1952); prey mobility may also be a factor in the

approach of Phidippus regius to prey (Edwards, 1975). Araneids possess

an innate (not learned) versatility in their handling of prey (Robinson

and Robinson, 1976), and it is likely that much of the versatility in

the approach of salticids to various types of prey is also subject to

genetic determination.

The experience of the spider may also be a critical determinant of

subsequent behavior. Salticids can learn to alter their response to a

particular type of prey for an indefinite period after a disagreeable

encounter (Drees, 1952). Gardner (1964, 1966) found that Phidippus are

more responsive to prey after longer periods of food deprivation.

Satiated Phidippus tend to lose their interest in the pursuit of prey.

Salticids will run only a limited number of pursuits in a given

situation, a result which has been ascribed to the fatigue of a parti-

cular hunting instinct (Drees, 1952; Precht and Freytag, 1958; Gardner,

1964; Dalwigk, 1973). However, Plett (1975) found that recovery to

this "fatigue" was stimulus-specific, and the waning performance of a

spider could be considered a temporary acceptance of the futility of a

particular form of action.

This interpretation of fatigue is consistent with the predatory

behavior of Phidippus observed in the laboratory. Individuals which

will no longer pursue a prey at the distance of 25 cm if this pursuit

would require a detour, will readily pursue the same prey at the same

distance if the route of access is direct. This applies to individuals

which had previously demonstrated an ability to negotiate the detoured

approach without any difficulty. Other individuals may not move from a

waiting position unless the prey remains in position, or immobile, for

a period which may exceed 15 sec.

Therefore, and admittedly in the absence of conclusive data, it is

proposed that the prey distance, the position of the prey relative to a

route of access visible to the spider, and the duration of prey

immobility are all factors which, in combination with the experience

and innate disposition of the individual spider, determine the proba-

bility of pursuit in a given situation. Much of the behavior of

Phidippus subsequent to the initial orientation turn to face a potential

prey can be interpreted, in a functional sense, as the evaluation of the

feasibility (or futility) of pursuit. When the prey appears in a

position which is readily accessible, the spider may pursue without

hesitation. As noted above, a lengthy waiting period may precede the

initiation of a detoured pursuit. This waiting period, although it

may be of use for the evaluation of potential routes (see below), may

also serve to measure the probability that the prey will still be in

its position when the spider, after some effort, has attained that

position. This would be true if the probability that a prey will

remain in position for a particular interval of time correlates posi-

tively with the probability that it will remain in position for the

subsequent interval. If prey tend to alight and then take off quickly,

a lengthy pursuit is wasted effort. This appraisal requires at least

a generalized ability on the part of the spider to assess the effort

required for the completion of a detour, an ability which is consistent

with the imperative role of vision in the selection of a detour by the


In a more general sense, this hypothetical versatility agrees with

the observed ability of Phidippus to conduct a directed search when

prey are sedentary, as mentioned earlier. It appears that these

spiders can adjust their behavior to account for the mobility of the

available prey, in determining an effective strategy of predation.


Salticids use vision to evaluate objectives and objective posi-

tions. In a corridor, salticids of the genera Evarcha, Heliophanus,

Marpissa, Salticus, Sitticus, and Synageles will move toward a striped

wall in preference to an unmarked wall (Kgstner, 1950; Dzimirski,

1959). In Figure 22 a similar demonstration of the role of vision in

determining a direction of movement by Phidippus is presented. One

significant feature of the ascent shown in this figure is the tendency

for this spider to stop and conduct an extensive visual survey whenever

the ascent is interrupted (decision points). The visual determination

of an objective, which in this case is clearly a vertical rod in the

context of ascent, consistently precedes movement in a horizontal plane

from each decision point. Salticids do not grope blindly in space when

visual information is available, but move under the direction of

specific objectives.

Figures 4 (B, E) and 23-26 (corresponding to data given in Tables

2-4) provide several different demonstrations of the role of vision in

the selection of a route of access to the prey position. As noted

earlier, the determination of the immediate direction of pursuit is

related in each case to the relative position of a plant configuration

which is part of the total route of access. This configuration is

termed the secondary objective, since it replaces the primary objective

(prey) as a determinant of immediate behavior.

10 cm

4 2
4 1/,

H l0 cm

S0 cmc


Figure 22. Sequential T-maze problem. The individual T, constructed
of wood dowels, is shown in (A): The spider ascends (1) to a decision
point (2), then either moves toward the visible route of ascent (3) or
in the opposite direction (4). The structure shown in both horizontal
plane projection (or view from above; B) and lateral perspective (C),
provides the ascending Phidippus with a series of 10 consecutive T
choices during each ascent. Each of the 10 decisions occurs at a level
10 cm above the level of the preceding decision. In a total of 10
ascents upon this structure, a single male P. pulcherrimus (SN 7mPUL)
moved in direction (3) as indicated in (A) in 97/100 choices (P-40.001).
In each trial this spider completed a steady walk to the top of the
structure. Pauses for visual survey always occurred at the decision
points, indicated by open circles above. The same individual completed
a run to the top of this structure in the dark (as observed by faint
red illumination from below), but in this case only 4/10 of its
decisions were directed toward the route of ascent, and there was no
pause for visual survey prior to a decision. In all cases the spider
walked on top of the horizontal bars.

5 cm

Figure 23. Demonstration of the role of vision in the determination of
a route of pursuit, irrespective of gravity. Each problem can be
repeated at a series of positions on an artificial plant constructed of
wood dowels. Both problems (A) and (B) begin with the presentation of
the standard lure (hanging dead fly) to the spider as shown (F). The
spider (open circle on main stem) immediately turns to face the fly
(1), to initiate the trial. It should be noted that the relative
situation of the prey in a horizontal direction perpendicular to the
vertical main stem on which the spider rests is the same in both
problems; the critical difference, then, is the relative position of
the physical connection (shaded) between the main stem and the position
of the fly. The spider is lured into an appropriate starting position
with the fly, prior to each trial. Pertinent data includes the number
of trials in which a distinct orientation toward the access route (2)
is observed, the number of trials in which the spider moves toward this
access route (3; this requires a descent in A, and an ascent in B), and
finally the number of trials in which the spider turns away from the
access route, to move either up (A) or down (B) the stem after sighting
the prey (4, above). If vision were a significant determinant of the
direction of pursuit, one would expect event (3) to occur significantly
more often than event (4), in both problems. This expectation agrees
with the actual results (Table 2).

Table 2. Results pertaining to the access problems (A) and (B) as
they are presented in Figure 23.


turn to
route (N2)

3.Run down
to route

IfPUL 30 19 29 1 40.001
3fPUL 22 16 20 2 40.001
5fPUL 20 2 20 0 <0.001
7fPUL 4 1 4 0 -
total PUL 76 38 73 3 <0.001
22fREG 40 31 33 7 <0.001
24fREG 15 14 15 0 <0.001

*In some cases, a direct jump down to the access route.


SN N 2.Distinct 3.Run up 4.Run down P(N3= N4)
turn to to route (N14)
route (N2) (N3 )

IfPUL 22 10 22 0 <0.001
3fPUL 32 13 32 0 <0.001
4fPUL 6 0 6 0 -
5fPUL 38 5 38 0 <0.001
6fPUL 22 3 22 0 <0.001
total PUL 120 31 120 0 <0.001
22fREG 44 1 44 0 <0.001
24fREG 64 3 64 0 <0.001

The numbered column headings (2-4) correspond to events described in
Figure 23. Both P. pulcherrimus (PUL) and P. regius (REG) were
observed. Clearly, as shown by chi2 analysis of these enumerative
data (column at far right), the decision to either ascend or descend is
based upon the relative position of the access route, which may be
termed a secondary objective. Although distinct turns toward the route
of access are not observed in most of these trials, prior to the
initiation of pursuit, this decision is clearly the result of visual
evaluation of the relative position of the access route; thus this
process is complete in most cases before the spider even turns away
from the prey, in this particular problem. In every case of ascent (4)
in problem (A), the spider moved up the stem to position for a direct
jump toward the prey.

4.Run up

P(N3= N4)


; 7


5 cm

C (SN 6fPUL)

B (SN 5fPUL)

D (SN 5fPUL)

Figure 24. Some typical solutions to the problems of access presented
in Figure 23. In each case the fly was quickly removed with no effect
on the spider as soon as the spider turned away from the initial
orientation to run a pursuit. Thus the subsequent pursuit is not
dependent upon a continuing prey stimulus. A: The spider turns to
face the prey (1), turns down to the access route (2), runs down on the
side of the stem facing the access route (3), turns (4), and jumps (5)
to the secondary objective, ascends (6), stops for a series of small
turns (7), ascends (8), and finally (9) explores the apical region
(green paper "''leaf"'') of the branch. B: Typically for this individual,
the initial orientation (1) is followed by a fast walk (2) directly
toward the access route, then ascent (3) and small turns (4). C: Turn
to prey (1), fast walk (2), halting descent (3), and interrupted small
turns (4). D: Turn to prey (1), slow walk on side of stem opposite to
the prey direction (2), reorientation (3), walk (4, 5), and small
turns (6).

5 cm


Figure 25. Demonstration of the visual determination of the access
route on a "circle of plants," as viewed from above. Each long
rectangle is a green paper "leaf," skewered (at center) by a long
vertical wood dowel (stem). Problems (A) and (B) are run alternately
for the most part. In each trial the spider first turns to face the
prey (standard lure; 1). As in the preceding experiment (Figure 23),
distinct turns toward the connecting segment of the access route (2),
and the subsequent direction of movement by the spider (3 or 4) are
recorded. The hypothesis of visual determination requires that event
(3), or movement toward the continuous access, occur significantly
more often than the alternative (4). Results which are consistent
with this hypothesis are presented in Table 3.

Table 3. Results for the problems of access presented in Figure 25.

turn to
route (N2)

access (N )

4 .Run
away from
access (N4)

IfPUL 24 22 24 0 O.001
3fPUL 37 24 37 0 -0.001
5fPUL 5 5 5 0
22fREG 70 45 60 10 <0.001
23fREG 42 3 41 1 <0.001

Since a comparable number of trials was run for each of the two mirror-
image problems described in Figure 25, with no significant difference
in the results, data for problems (A) and (B) are pooled in this table.
The numbered column headings correspond to events described in Figure
25. Individuals of both species (P. pulcherrimus and P. regius)
exhibit a highly significant tendency to move in the direction of the
visible route of access. In this situation, most individuals make a
distinct turn (2) toward the connecting route (or plant) before
turning to run the pursuit, in a majority of their trials.

P(N3= N4)


\ t

3 /
4 3


4 33O^^ e

Figure 26. Problem of access requiring the use of a connection which
is removed from the plane defined by the immediate route and the prey
position. This problem is shown in perspective (A), and also as both
left- (B) and right-handed (C) configurations, as viewed from above.
As with the preceding problems, turns directed toward the connecting
route (2), as well as the direction of movement (3 or 4) from the
starting position, after the initial orientation to the prey (1), are
recorded for each trial. Since data obtained for either of the two
possible configurations of this problem are comparable for each of the
individual spiders, these data are pooled in Table 4.

Table 4. Results for the problem of access in three dimensions which
is described in Figure 26.

SN N 2.Distinct 3.Run 4.Run P(N3 = N4)
turn to toward away from
route (N2) access (N ) access (N4)

1fPUL 41 41 41 0 <0.001
3fPUL 40 40 40 0 <0.001
22fREG 38 30 34 4 <0.001
25mREG 40 34 35 5 <0.001

The numbered column headings correspond to events described in Figure
26. Each spider has not only a highly significant tendency to move in
the direction determined by the connecting route of access, but also a
significant tendency to turn directly toward that connection most of
the time before pursuit. Each turn toward the secondary objective (2)
is quite distinct. The two female P. pulcherrimus repeatedly ran the
problem in either direction like clockwork, never missing a distinct
turn toward the access route prior to pursuit. This contrasts somewhat
with the results of the two preceding experiments, where such turns
were often omitted by the same individuals. In this problem, however,
the connection between the access route and the immediate route is more
complicated. All of the features relevant to the solution of this
problem are not coplanar.

In a substantial majority of the access problems presented to

various Phidippus during the course of this study, the spiders almost

invariably utilized the most direct route of access to the prey that

was available. One might think that a considerable amount of insight

is required to acheive this particular result, but it is actually

easier to explain this efficiency with a simple evaluation of the

constraints in the spider's choice of a direction of movement upon the

plant. A stem, for example, allows for only 2 directions of movement.

A simple rule for behavior in choosing a direction of movement on a

stem during the pursuit of prey can be stated: 'Move in a direction

which forms an acute angle with the direction of the prey. This basic

rule is an effective predictor of the behavior of Phidippus on a

simple, isolated runway in a horizontal plane; here compliance with

the rule invariably leads the spider to the closest possible approach

to that prey. Nonetheless, this "''rule"'' is clearly violated by the

spider in many situations, including the problems of access shown in

Figures 23 and 26.

These apparent violations of the acute angle rule can be recon-

ciled with that rule, given a new understanding of the role of the

secondary objective. Any objective (primary or secondary) can be

viewed as a determinant of the behavior of the spider with respect to

the choices offered by the immediate route, in a manner consistent with

the acute angle rule given above. During each segment of pursuit, the

spider is in compliance with the acute angle rule in moving toward its

immediate objective; the rule can be stated in a more general form:

Move in that direction (afforded by the immediate route) which forms an

acute angle with the direction of the immediate objective. In this

form the rule actually provides a functional definition of the term

objective. The ability of Phidippus to solve a complicated problem of

access is dependent upon its ability to substitute intervening, or

secondary, objectives for the primary objective, as determinants of

immediate behavior.

A fundamental question, then, is the nature of the visual

relationship between the prey position and another position in space,

which determines the role of the latter as a secondary objective.

Clearly the secondary objective should be selected on the basis of its

ability to afford access to the prey position. With a slight rephras-

ing, the acute angle rule can also be applied with some predictive

value here: Provided that it is not too far away, move toward a

secondary objective which is situated in a direction which forms an

acute angle with the direction of the prey. Then, if direct movement

toward the secondary objective is not possible, the acute angle rule

can be applied to predict the direction of movement on the immediate

route toward the secondary objective. If several potential secondary

objectives are available, then the spider might simply move toward the

first sighted position which meets certain criteria.

Certainly the pursuit behavior of Phidippus involves a more versa-

tile approach to the problem of access than one can describe with a

simple rule. As shown in Figure 27, the route of access which is

employed by Phidippus does not necessarily consist of a visible plant

configuration. Jumping routes and lines of silk can also be utilized.

Certain features of individual behavior are best explained by their

contribution to the concealed approach of the spider. As noted by

Holling (1966), a direct route selected for purposes of speed may



\ \ /5 cm


Figure 27. U-e of a dragline and a jump to complete a route of access,
by a Female P. pul cherrimus. After the initial turn to face the prey
(1), the prey '-standard lure) is immediately removed. Subsequently,
the spider locates and climbs under an existing dragline (2), reorients
(3), turns down toward the plant associated with the primary objective
(4), jumps down (5) and recovers (6) to ascend the objective plant in a
series of short runs (7).

conflict with the advantages of concealment during an approach to a

particular type of prey. As shown in Figure 24 (D), certain spiders

would consistently climb up the opposite side of the stem from the prey

direction, during each pursuit which required an ascent. Many indi-

viduals only behaved in this manner on occasion. The fact that this

behavior is part of the repertoire of at least P. clarus and

P. pulcherrimus, once again suggests the versatility of Phidippus in

selecting a strategy which may be appropriate for certain prey under

certain conditions. Phidippus consistently descend a vertical stem on

the side facing either the primary or the secondary objective.

In summary, Phidippus can set a series of immediate objectives,

beginning with the position of the prey (primary objective), during the

course of pursuit. Selection of secondary objectives presumably

relates to their position with regard to the primary objective, in

terms of access to the primary objective. This selection can be the

result of visual evaluation of the problem of access by the spider.

The behavior of the spider on its immediate route is consistent with

the pursuit of the immediate objective; by moving in a direction which

forms an acute angle with the direction of the immediate objective, the

spider approaches that objective as directly as possible. The use of

a series of secondary objectives can lead the spider to utilize the

most direct route of access that is available in many situations.

As noted earlier, the additional ability to maintain a memory of

the relative position of the prey at all times during pursuit is as

essential as the ability to pursue appropriate secondary objectives.

A long pursuit may consist of a series of segments, each initiated by

a reorientation to the expected position of the prey. After each


reorientation to the prey, the spider may select a new secondary objec-

tive, or immediate destination. Thus the choice of an appropriate

secondary objective during each segment of pursuit depends ultimately

upon the accuracy of the spider's ability to remember where the prey is



Of all the potential references which might be employed by

Phidippus in the determination of a direction in space, the orientation

of the body axis itself is the most immediate. In orienting toward

sighted prey in the absence of visual feedback, as described above,

these spiders are in effect utilizing the orientation of the prosoma

in space (corresponding to the positions of the eyes) as a reference

direction for the execution of an appropriate directed turn. In a

broader sense, the immediate route of the spider can also provide a

very useful reference direction during the pursuit of prey. Basic

terms which are used in the description of a simple segment of pursuit

are presented in Figure 28. Angles which specify a particular

direction with reference to the direction of a route are denoted as e;

hence the orientation of these animals with reference to the direction

of the immediate route is termed 8 orientation.

Figures 29 and 30 provide a description of the apparatus used to

measure the ability of Phidippus to utilize e orientation during a

segment of pursuit on a horizontal route. As shown in Figure 30, this

apparatus allows for the measurement of orientation and reorientation

angles in a horizontal plane. The arrangement and conduct of this

experiment effectively eliminate most visual cues which might be

employed in the determination of a direction of reorientation by the

Figure 28. Direct (A), symbolic (B), and analytical (C) diagrams of a
segment of pursuit on a horizontal bar. A, B: The spider turns to
face the prey (1), turns away to run a segment of pursuit (2), and then
reorients (3). C: Definition of terms used to describe this behavior.
e is the angle between the original spider-to-prey vector (magnitude D)
and the pursuit vector (magnitude S). e is the observed reorientation
angle with reference to the direction of pursuit; 9 is the calculated
value of e which would bring the spider to face the original prey
position. 9 is a function of 9, S, and D. H is the distance
estimate of the spider which is implicit in the determination of 9 ; H
is a function of the measured values of 8, S, and 9 L is the r
distance between the prey and the route of access.

Figure 29. Vertical section of apparatus used to measure pursuit
behavior on the horizontal bar, with some added perspective. The spider
(open circle) runs on the horizontal bar (HB) at center, a wood dowel
9 mm in diameter. The bar is surrounded by an inner (1) and an outer
(0) cylinder of heavy white paper. A lever (L) permits rotation of the
bar in a horizontal plane. The standard lure (F) can be dropped into
a trough between the two cylinders, out of sight of the spider, at the
beginning of each pursuit. A hanging incandescent lamp is centered
above the apparatus. The top of the horizontal bar is level with the
top of the inner cylinder.

Figure 30. Horizontal bar within inner (I) and outer (0) white paper
cylinders, as viewed from above (see also Figure 29). Prior to each
run, the spider is led back to a center position (1) on the horizontal
bar with the standard lure, to maintain a constant prey distance (D)
at the initial sighting. The prey (standard lure) is presented to the
spider in a circumferential position (2). Subsequently, as the prey is
dropped out of sight of the spider, the spider runs to a new position
on the horizontal bar (3) and reorients in the direction of position
(4) on the circumferential (F) scale. From values 1-4, 8, S, and e
are determined. e is also calculated for each pursuit, as it is r
defined in Figure 28. Running (R) and circumferential (F) scales are
divided into 1 cm and 3 units respectively, as shown above. For each
pursuit the values 1-4 are read in sequence by an observer looking down
from above. The accurate determination of (4) requires the most
critical attention on the part of the observer.

spider, apart from those provided by the route itself. In addition,

gravity is of no use for the determination of a direction in the

horizontal plane.

The results of this experiment are given in Figures 31-38. For

each series of trials, involving the presentation of prey at varying

directions with reference to the horizontal bar (variable orientation

angle, 9), the measured angle of reorientation with reference to the

route (6r) is presented first as a function of 0 (Figures 31, 33, 35,

and 37), and then as a function of 9c (Figures 32, 34, 36, and 38), as

the latter is defined in Figure 28. From this presentation it is

possible to draw several conclusions. First, for each of the indi-

vidual spiders (including P. pulcherrimus, P. clarus, and P. princeps),

the actual reorientation angle (6er) correlates quite significantly with,

and approximates somewhat, the initial orientation angle (6). 8r is

clearly a function of 6.

In addition, 6 tends to be significantly greater than 9 in
magnitude. This is demonstrated by the fact that, for each spider,

r 8 is significantly greater than 0. The graphic presentation of 9r

as a function of 9 (Figures 31, 33, 35, and 37) provides a clear

exhibition of this fundamental result. From an examination of Figure

28 (C), one can see that the expected value of 9r, if it were compen-

sated for the relative movement of the spider, should exceed the value

of 0. As the spider approaches the prey (Figure 28), the reorientation

angle required to face the prey position directly (6c) increases. From

the data presented in Figures 32, 34, 36, and 38, it is evident that 6c

is an effective predictor of the measured reorientation angle (er).


I I .... I I ....... I i '
SN 3fPUL /
90 /0

/,:*>* . %
t**:! t -

*! r/ -- t
0 6 0 *b e b e lb

U | !o, /-<***f
^ :* : ;**: : A
6 *s 00 W"

^ *' .:.' vS 'X
H- a ^
'se / e 0
j N= 322
30 r= 0.89
o 30 z test:
:P(r= 0)4 0.0001
,* S-= 4.4 cm
/' * e -e= 11.4 + 0.5
/ / r --'
'' 11i' t test:
P(9 = 0) 0.0001

0 30 60 90


Figure 31. Behavior of a female P. pulcherrimus on the horizontal bar:
the measured reorientation angle with reference to the direction of
pursuit (9 ) as a function of the initial orientation angle (9). There
is no doub[ that the magnitude of the reorientation angle is strongly
correlated with the magnitude of the orientation angle; 9 is also
significantly larger than 9. As a standard procedure in [his and subse-
quent figures, the linear regression of Y (e) on X (9) is indicated as
a dashed line.

.. '.. L
90. /-0

Ou 60 0
....:.. .;.- y
: :::**:'y
*O* 9:* '

6O 4.06 : 0*e 1 -
60 : ; : "


M z et
seeP seOf< 0.0
g .'~*.I- -C*

S-- a:*^ %

0 30 N= 322
-6il. 0 =O94
h- l r s (/ l
'/0k 0 r=?

( o/ie o *a z test:
.i"' P(r= O) '//0:. S-= 4.4 cm
*^ *^
,,)- -

0 I I I I I ]
O 30 60 90


Figure 32. Behavior of a female P. puicherrimus on the horizontal bar:
te as a function of e^, same data shown in Figure 31 As indicated by
the linear regression (dashed line), 9c is a good predictor of the
measured values of er. As shown in Figure 28 (C), e9 is the reorien-
tation angle which would bring the spider to face the original prey
position directly after movement of magnitude S; it is a function of 8
(initial orientation angle), S, and D (initial prey distance).

90 -

/ /
*: ^

03. 0
% 0
o 9 /
*0 000** /*0 0
-~~ .^ *r-
60 0-. 0 *'
.u .. . 4 2 -

C) 0 ;IIA 0
H- ~ /00^ 0/ 0
Z N= 240
/ S

S~*r. r0.85
o 30 A /6 z test:
go. P(r= 0)<0.0001
/.,S; U= S4.5 cm
/0 8* r-6= 11.2 + 0.7
o )/ t test:
-/ P( -6= O) r

0 A
0 30 60 90


Figure 33. Behavior of a female P. pulcherrimus on the horizontal bar:
Or as a function of 8. As for the data presented in Figure 31, the
correlation of 9 and 9r is highly significant, and Or is significantly
greater than a corresponding 6.

L. 04 *. 9W0
I o ,%. * I _"^ ,s
to \ ::
60 o ." -

0 .
. .. : -
< **Sl ^**
< .: /

"r= 0.88
< z test:

do o (r= 040.0001
H- *. ,'* *

,5T= P 0cm
LU,. : c
0 / *
o / 0 i

0 30 6o 900

Figure 34. Behavior of a female P. pulcherrimus on the horizontal bar:240

8,as a function of 8c. The analysis of these data is consistent with
H-at presented in Figure 32. r=.
u ~/
/^*' ,. z test;
/A /4 P(r= 0)< 0.0001
*/s 'S= 4. 5cm


0 30 6090


Figure 3J^i Behavior of a female P. pulcherrimus on the horizontal bar:
e as a function of 9^. The analysis of these data is consistent with
that presented in Figure 32.

gO /0

S 0 /t o o

S0 o / 60

Q0 o/

g ee/ 0 0
0~^- / / *
S .!: *,*'

0 / 0
^ ^ :/" *:
*@00- l

.4 0 '
r 0 / -.

Z / / /
-- ",/ N= 148
C o-/ e/ r= 0.90
0 / z test:
SPr= 0) <0.0001
/ / = 4.8 cm
6/ --e= 12.8 + 0.8
?.. 0 /

7 r
/. o" t test:
*/ P(r-= 0) 0.0001
,/ r

0 30 60 90


Figure 35. Behavior of a male P. clarus on the horizontal bar: 9 as
a function of 8. Analysis of these data is consistent with that
presented in Figure 31.

900. ./ cJ

Owe s / -
i*J 6 4 /
i** 0 ,
CD. *.3*3<**

60 0/0f
z 6o

o /0/ /,
.._ .. / /
< 3
I- 00. -

I-- l
U ^/ 0t0.

o / = 4
L0 0 S
/ *ztes/ /t

"I 1 i
S303 /
-, ^0 N= 148


Figure 36. Behavior of a male P. clarus on the horizontal bar: 0 as
a function of 8 The analysis of these data is consistent with t~at
presented in Figure 32.test:
^/- P(r= 0)<0.0001
0/"*. S= 4.8 cm
'/ ge.

0 30 60 9Q0


Figure 36. Behavior of a male P. clarus on the horizontal bar: e as
a function of e The analysis of these data is consistent with that
presented in Figure 32.

900 ......... 1 ,I- ,
SN 19fPRI (o) o
N= 137
r= 0.78
T= 4.1 cm 09 o *
0= 9. 1 + 0.7 0
0 0
60 o o
a :00 080
.-,o *o 8 tf>
o o o | $

__ 0 Co / SN 20mPRI (o)
_\ */ 0 N= 84
< r= 0.86
| S'S 2.9 cm
<- -r-=4.4+ 0.8
So30 r -

19 0 Both spiders:
'a P(r= 0)< 0.0001
Sr(e -e= O)< 0.0001

0 r

horizontal bar: e as a function of 8. These data were collected on
the modified horizontal bar as shown in Figures 48 and 49. The
numbers (19) and (20) demarcate the linear regressions for the two
spiders, by number. Analysis of these data is consistent with that
presented in Figure 31.

9 0 0 1 1 1 ... i I 1,1-1 I

SN 19fPRI (.)
N= 137 *
r= 0.85
z test: <
P(r= 0)< 0.0001 o %0

| 60 o 0

0 8 0
z o * a

.0 60 0 S
H- p *

< 30 1*0
F- '*W8%

S-* *tv p O

3 SN 20mPRI (o)
E00 N= 84
< r= 0.90

19 z test:
P(r= 0)< 0.0001
/S= 2.9 cm
/ /20
I I .3-- -
0 30 60 90


Figure 38. Behavior of a male and a female P. princeps on the
horizontal bar: er as a function of 6c. The numbers (19) and (20)
demarcate the linear regressions for the two spiders, by number.
Analysis of these data is consistent with that presented in Figure 32.

It is evident that each reorientation turn represents an attempt

by the spider to face the expected position of the prey. Within the

context of some reference system, Phidippus can definitely remember

the relative direction of the prey, as shown by the accuracy of the

reorientation turn. The close relationship between the measured value

of 9r and the calculated 6c for a given pursuit suggests that, if the

spider is actually using the direction of movement as a reference

direction (memory constitutes use of e information), then it may also

be able to compensate for its own displacement during pursuit (of

magnitude S) in determining an appropriate reorientation [since c=

f(e, S, D)]. Therefore a more rigorous demonstration of the role of

the route, or the immediate direction of movement, in determining a

direction of reorientation would be useful.

Such a demonstration is provided in Figures 39-41. When the

horizontal bar is rotated by 900 during the course of pursuit, the

spider still reorients according to the reference direction provided

by the direction of its movement upon the bar, irrespective of the

rotation of the bar, and apparently in preference to conflicting data

from background visual cues available in an open room. In addition, as

shown by the data presented in Figures 42 and 43, this reorientation

behavior (with ec as an accurate predictor of 9r) does not depend on

any visual feedback in its execution. After the initial orientation to

prey, Phidippus can complete a segment of pursuit, followed by an

accurate reorientation turn, in complete darkness.

As shown in Figure 44, even second reorientations in response to a

single stimulus follow c, or 6, with an appreciable accuracy.

A -' II




Figure 39. The horizontal bar before (A) and after (B) a +90 rotation
in the horizontal plane, as viewed from above. Rotation takes place
during the pursuit run of magnitude S. In (B), dashed vectors indicate
two possible directions of reorientation. 8e is an angle of reorien-
tation (measured with reference to the bar) which is predicted by the
spider's use of fixed, external cues to determine the direction of prey.
Consistent with earlier definition, ec is the angle of reorientation
which would be expected if the spider utilized its direction of move-
ment upon the horizontal bar as a reference direction to determine the
direction of prey. The outer cylinder of this apparatus (0 in Figures
29 and 30) is removed for this rotation experiment; thus visual
patterns within the open room are available for reference as an alter-
native to the use of the horizontal bar, which is rotated. Nonetheless
measured reorientation angles with reference to the route (er)
approximate ec rather than ee, as shown in (B) above, in support of the
general hypothesis that the direction of movement upon the immediate
route provides the reference direction in this situation. Data is
presented in Figures 40 and 41.


900 SN 3fPUL /

J ///
-6 / t
o o o

-J **/ *

< 60. 0 *y

< / S
o /
i- oO '
S/ 9 /

-/ o

S30 N= 84
S/ r= 0.68
< / / z test:
/ / P(r= 0)< 0.0001
I / S- .
/ / 67-= 7.1 + 1.60
/ / S

I I 'l [ I I I I Il I I I

0 30 60 90


Figure 40. Reorientation by a female P. pulcherrimus after a 90
rotation of the horizontal bar during pursuit, as described in Figure
39. In every trial the spider directed its reorientation toward the
same side (right or left) of the bar as its initial orientation to the
prey. Regardless of the rotation of the bar in an open room, ec
remains a highly significant, and accurate, predictor of er. As shown
in Figure 39, an alternative direction of reorientation based upon the
use of a fixed reference system (8e) deviates from the direction speci-
fied by ec by 90; use of the fixed reference system in this situation
would lead the spider to reorient toward the opposite side of the bar
from the direction of the initial orientation. This has not been
observed. As for the data presented in Figure 41, roughly half of the
trials presented above correspond.to the "left-handed" problem shown in
Figure 39; the remaining trials represent the results of the similar
right-handed problem involving presentation of the prey to the right
of the direction of pursuit and a rotation of -900 during pursuit.


900 -

: A^
*L < A
D 0 0 0 0 0
ng *. O A
** A^

C) 0
S60 -~ Ai-3, 6
D *
I- A .*

< 30
F-- -

< "N= 107
J r= 0.52
A 0 0
0 0A1

z test:
P(r= O)< 0.0001
LU A' /
A o

S,- = 4.0 + 107
'^ / ,r= 0.52
/ z test:
/* P(r= 0) / ^ 9-6 = 4.0 + 1.3
/ *< ~r c -
0 1iI .." I I '
0 30 60 90


Figure 41. Reorientation by a male P. pulcherrimus after a 90
rotation of the horizontal bar during pursuit, as described in Figure
39. As for the data shown in Figure 40, every reorientation turn was
directed toward the same side (right or left) of the bar as the initial
orientation to prey; 8c is a highly significant predictor of 8r


90 SN HfPUL //

j 00 /
/ /

0 /%

C 60 r* 0. 8*

A z test:**
_.j **r

0 Pr 0.< .000
/* 90/

I- S/ -em
2: / /
-- /4 N=120
o 30 // .o r= 0.78
/ *I// z test:
"/1ie" P(r= 0)<0.O00l
S= 6.8 cm
/ e r-e= 12.6 + 0.9
-/ t test:
P(6r-9= o)< 0.0001
0 I I
0 30 60 90


Figure 42. Reorientation turns executed on the horizontal bar in
complete darkness by a female P. pulcherrimus: 8r as a function of B.
Using the apparatus described in Figures 29 and 30, the overhead light
was switched off as soon as the spider turned to run in pursuit. Only
those trials in which the reorientation turn had been executed
completely prior to the time at which the light was subsequently turned
on (several seconds later) are shown above. In many trials, this
spider did not reorient until the light was switched on, and these
trials were not recorded. In one of the trials shown above, the spider
was even preparing to jump in an appropriate (prey) direction as the
light went on. In all respects these data compare with those collected
under conditions of continuous illumination. Completion of this
experiment was facilitated by the tendency of this individual to make
rather long runs (S= 6.8 cm) prior to reorientation.


... I| I I...I I I,

0 SN I1fPUL *I ,'
9 /


C //
/ .

o 6o.-
0 -0
o *
F- a^ *
2: @1 a -

w 30 /.*
<~> 0* 0 *


N= 120
,' r= 0.83
/ z test:
7 P(r= 0)< 0.0001
/ S= 6.8 cm
^/ 0
030 60 90


Figure 43. Reorientation turns executed on the horizontal bar in
complete darkness by a female P. pulcherrimus: 8 as a function of 8c.
Even when the reorientation turn is executed in the dark, 9 is
an accurate predictor of its magnitude. The trials shown above corres-
spond to those presented in Figure 42.


90 - -,

SN 5fPUL o ///
/ ,

0 /00

80 */00
00 / /
608 a,//" -,0

S0 0 /6 /0/
So o o.O.; rR
-- 0 FIRST TURNS (.)
*o N= 60
// // *r= 0.94
/ r' S= 8.0 cm
0; / J3 */O
30 0
<- / / o SECOND TURNS (o)
S* /o N= 28
< r= 0.83
/ S= 10.4 cm
/ / .

y/0. 0

0 30 60 90


Figure 44. First and second reorientation turns executed by a female
P. pulcherrimus in response to a single stimulus (for each trial).
The accuracy of a second reorientation turn compares favorably with
that of the first reorientation. For both turns, P(r= 0)4 0.0001.
Numbers (1) and (2) demarcate the linear regressions (dashed lines)
corresponding to the first and second turns, respectively.

Clearly, then, the angle between the initial orientation to prey

and the direction of movement (e) is a significant determinant of the

reorientation angle with reference to the direction of movement (Or)

In addition, e is a significant determinant of S, the distance of the

pursuit prior to reorientation (Figures 45-47). With reference to the

definitions of ec, S, and L provided in Figure 28 (C), the following

relationship exists: 2
dec sin ec

dS L
Since sin c increases to 1.0 as ec approaches 90, one can conclude

that the instantaneous rate of change of the angle between the route

and the direction of the prey, with respect to the distance moved by

the spider during pursuit (d9c/dS) also increases as ec (the angle

between the route and the direction of prey) approaches 90 given a

fixed value of L (distance between the prey and the route). As shown

in Figure 45, the hypothesis that a fixed value of (ec-9), or A8c,

determines the length of pursuit prior to reorientation (S) is consist-

ent with the central tendency of the observed behavior (allowing for

considerable variation around this central tendency). A rationale for

this observed tendency can be developed as follows: When 9c is small,

an accurate reorientation is possible after a longer pursuit (distance

S), since ec is changing so slowly with respect to S. For a given prey

distance (D), a longer run is-also required to bring the spider to the

closest approach possible on a particular linear route, when c

(initially equal to e) is small. Conversely, when 9c is larger

(approaches 90 ), more compensation for movement relative to the prey

is required in the determination of a reorientation direction, due to

15 cm t-. ,-

N= 322
r= -0.73
z test:
S\P(r= 0)4 0.0001
10 \ D= 25 cm (constant)

-13 q .,:.
. \,
U* 5

5 ">-- \ :\
z I- --
S* *. 1 ..

| * A;*^^ ~ -^ -- ^
S* I 3.L Id-2Lki.L.----
-- \ "--t" ^.
SN 3fPUL ++k '-
X \
0 1 ------1--1-- i ------.-- i1 ---- 0
0 30o 60 90


Figure 45. Behavior of a female P. pulcherrimus on the horizontal bar:
S as a function of 8. Data from this set of trials are also shown in
Figures 31 and 32. This spider exhibits a highly significant tendency
to run a greater distance when the prey appears at a smaller angle with
reference to the direction of movement. Both linear regressions of S
on e (Y) and 8 on S (X) are shown above as dashed lines. Curves (A) and
(B) are based upon the equation [S= H(cos 8-sin 8/tan 9c)], where ec=
(8+110) and H is equal to either 20 cm (curve A), or 15 cm (curve B).
If the position at which the spider stops to reorient were determined
by a change in the respective direction of the prey (indicated by 9c
with reference to the direction of movement) of a fixed magnitude (in
this case 110, corresponding to a central tendency shown in Figure 31),
then, depending upon actual H values ("distance estimates" as defined
in Figure 28 (C); values used to define the curves above are consistent
with data presented in Figures 55-57), curves such as (A) and (B) above
could be expected to describe the relationship of S and 8 with some
accuracy. Although, as indicated above, 8 is a highly significant
determinant of S, the latter also varies greatly for any given value of

\\x X

10 1t 0 0 a

$ \
*i \
* \

N= 240
r= -0.52
z test:
P(r= O) 0.0001
D= 25 cm (constant)

* 0 S

0 : &

.0 "\ t \.

00 0.0 0 a
* *to I *'s
......+.+** }* *,-+ *-.
* ** \SbA

"* '| *i \s "

: 6 0
*50 go

~ I :

* .:! 1\\40 :

- Y



Figure 46. Behavior of a female P. pulcherrimus on the horizontal bar:
S as a function of e. As in Figure 45, both linear regressions are
shown. Although e is not as precise a determinant of S for this indi-
vidual, a highly significant negative correlation does exist between
the two values.

15 cm

5" "



~. .

. \

S N., 5

\ .

N= 148
r= -0.63
z test:
P(r= 0).- 0.0001

D= 25 cm (constant)

* 0 9


SN 2(

* 0 0

0 \\
: -* :^* : :-*

2 : { *I ", ***.

6mCLA S + N


b636 '6b g0o0


Figure 47. Behavior of a male P. clarus on the horizontal bar: S as a
function of e. Both linear regressions are indicated as dashed lines.

15 cm

* *\

10 *


J m I

I J I I m


a higher rate of change of 8 with respect to S when the former is
large (near 900). A decision to either jump or change to a new second-

ary objective is more imminent when the spider is nearer to the closest

approach to prey on a particular linear route (where e = 900). Regard-
less of which rationale for the adaptive value (function) of this

behavior is appropriate, it is clear that e can play a significant role

in the determination of S. As e increases, the observed value of S

tends to decrease (Figures 45-47); nonetheless it is equally clear that,

given a particular value of 9, S varies greatly.

The determination of an accurate reorientation angle (a value of

9 which approximates the calculated 6 ) involves more than an ability
r c

of the spider to employ a memory of 9, as shown above. Based upon

definitions provided in Figure 28 (C), one can conclude that 9 is a
function of D (the initial distance of the prey), e (the initial

direction of the prey with reference to the route of access), and S

(the distance of movement along that route prior to reorientation):

tan-l F sin 8 1
C cos 6 (S/D)

By this analysis, the spider should be able to integrate these 3 forms

of information (D, 9, and S) in developing an accurate appraisal of the

relative direction of the prey with reference to the route (8 r).

It is possible that the fact that 9 is significantly greater than
9 could represent the use of a simple adaptive output by the spider,

based upon an ability to remember 9 combined with the neurological

analog to the generalization that 9 is always greater than 9 to some
extent. This might in itself provide a reorientation angle of suffi-

cient accuracy in most cases, independent of an ability to evaluate S.

For this reason, the modified horizontal bar described in Figures 48

and 49 was employed to measure the role of S as a determinant of e
when 8 is a constant (35 ). As shown graphically in Figures 50-54, a

highly significant positive correlation between S and r as predicted

by the equation given above, does in fact exist. In this experiment,

variation in S is an intrinsic feature of the behavior of each indi-

vidual Phidippus, when presented repeatedly with the same problem of


It appears, then, that Phidippus can utilize a memory of the

extent of their own movement (s) during pursuit to determine an

appropriate 9r. Previously, the ability to retain a functional memory

of distance travelled has been demonstrated for certain Hymenoptera

(Apis mellifera: Frisch and Jander, 1957; Gould, 1975; Cataglyphis

bicolor: Burkhalter, 1972; Duelli, 1976). Apparently certain spiders,

including the agelenid Agelena labyrinthica (Dornfeldt, 1975b) and the

ctenid Cupiennius sale (Barth, 1976) similarly retain information with

reference to the distance travelled (translation) during a relatively

short foray in pursuit of prey. It is possible that the dragline is

employed by these spiders in the measurement of route length, although

this is not known for certain. Subsequent to a presumed evaluation of

prey distance by Phidippus, a fixed length of silk line is played out

during the predatory jump (see Figure 61, below). In this case, at

least, the amount of silk played out should correspond to an estimate

of the prey distance. A Phidippus may, conversely, be able to deter-

mine the extent of its own movement by monitoring the quantity of silk

which is played out during that movement. Other methods, including the

r /
F ,V

Figure 48. Perspective drawing
positions are read from the (R)
the horizontal plane are read f
horizontal bar is 70 cm long; e
sists of a 100 cm long screen o

of horizontal bar in corridor. Spider
scale on the bar, and orientations in
rom an (F) scale, as shown above. The
ach of the walls of the corridor con-
f heavy white paper. See also Figure 49.

/ C\
// --- --- -- -- --

B 2__________, .4
//~ ~ ~ ~ 2. '345/ 8 02/22321425_ t262728 131 32 33 .4 5%3


J-T/ _r/ i? I r/b 1 617\/FI ,?\/yZ312] I^(z1\ 7W 20 I 321 i~-L 134 Ps^

Figure /49. Perspective view (A) and horizontal plane projection (B)
of horizonalI bar in corridor (also shown in Figure 48). A: After
the spider (open circle on scale R) sights prey (solid circle at left)
at an angle R (measured in the horizontal plane), the prey can be
dropped into a trough as shown at left, and thereby concealed from the
spider during thie pursuit and subsequent reorientation. B: The value
of e, S, and 01. are determined from recorded values 1-4, taken from
the R and F scales respectively, as shown above. By presenting the
prey to the spider at a constant position relative to the spider posi-
tion in each trial, 0 can be held constant (near 35), regardless of
the fposilt irn of the spider. Given this fixed value of e, the prey
distance ()) is varied by placing the screen (F scale, above) at
different di -.tainces (L) From the horizontal bar (R scale).

H= 12.8 cm ,
1J //
I: r --^,/ y
260 4 *
/ .- N= 95
A r= 0.91
z test:
(N= 61) P(r= 0)<0.0001

30 ]

E= 350 /
D= 19.4 cm N) N'C) "
H= 19.2 cm N
L- -/** ^ ^ / --/'* /

-1 0
< 60 -" *
0 /
CD & N= 109
B r= 0.85
/ )z test:
( ( N= 88) P(r= 0)<0.0001

30 -
0 5 10 15 cm
Figure 50. Behavior of a female P. pulcherrimus on the horizontal bar
in response to a constant 8 (350), for prey distance (D)= 13.6 cm (A),
and D= 19.4 cm (B): er as a function of S. Four reference curves are
defined by the function [0 = tan- (sin e/cos e -S/H)] for each graph,
based upon values of H (cm) which are given beside each of the curves.
In this, and subsequent figures, an arrow is used to demarcate the
curve which corresponds with an accurate determination of D by the
spider (as predicted by 0c; for points on this line Or= 6c and H= D).
For subsequent comparison (Figures 55-57), H values are calculated for
runs of at least 4 cm, and 9r of at least 40, as delimited in each
graph by the inset rectangles. H determinations from shorter runs are
considered to have less accuracy. Values of N given in parentheses
pertain to the number of trials which are thereby employed in the
determination of H.



A <




B -



0 5 10 15 cm


Figure 51.
in response
(B): Or as

Behavior of a
to a constant
a function of

male P. pulcherrimus on the horizontal bar
e (350), for D= 13.6 cm (A) and D= 19.4 cm
S. See Figure 50 for interpretation.


_L D= 19.4 cm /
H= 10.8 cm
LU 9 l- -? / / */

.160 N= 105
z 1 Tr= 0.56
-z test:
< .2 P(r= 0)40.0001
^/^^ '^,^^ -(N= 75)
A l 40-

SN 15mPUL Lo 0-
e= 35 Vo V
90 oD= 24.4 cm * 6 o O
H= 12.2 cm

6 06 09 60 go0909
0 S. *-S S.* 95 */

0 "" 9/ 0 go
LU +

< 0909 N=11]
S60 -/ -- /
z0 /r= 0.74
H- z test:
-P(r= 0)< 0.0001
LU / ^ ^ -^^
B N(N= 102)
oJ /
,' ',-- (N 102). .

0 5 10 15 20 cm


Figure 52. Behavior of a male P. pulcherrimus on the horizontal bar in
response to a constant e (350), for D= 19.4 cm (A) and D= 24.4 cm (B):
Sr as a function of S. For interpretation see Figure 50. Although the
correlation of S and er is highly significant for this individual, H
values are significant underestimates of D in each case.

L- _D= 13.6 cm C/"
S H= 19.3 cm / -
0 60
< /* - - ^

^ 60 4& /,. -' ,
A r= 0.59
( N- z test:
S/ / I* (N= 96) P(r= 0)<0.0001
0 /./ I, ,*^- -- -

^a^--* -0
.//- I;

e-= 35 I / N" /\
o-D= 19.4 cm C O / \-", <-"
LU H= 22.8 cm
I/ / / ".-J

B = m 9" i^.
D I 'Z- L > N= 10
B r= 0.63
~ / ^--- .- ~ z test:
(N= 102) P(r= 0)< 0.0001

30 -

0 5 10 15
Figure 53. Behavior of a female P. regius on the horizontal bar in
response to a constant e (35), for D= 13.6 cm (A) and D= 19.4 cm (B):
er as a function of S. Interpretation of curves and treatment of H
determinations is consistent with Figure 50.