Title: pollination of avocados
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027122/00001
 Material Information
Title: pollination of avocados
Physical Description: Book
Creator: Stout, Arlow Burdette
Publisher: University of Florida Agricultural Experiment Station
Publication Date: 1933
Copyright Date: 1933
 Record Information
Bibliographic ID: UF00027122
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aen4723 - LTUF
18204847 - OCLC
000924118 - AlephBibNum

Table of Contents
        Historic note
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        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
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
Full Text


The publications in this collection do
not reflect current scientific knowledge
or recommendations. These texts
represent the historic publishing
record of the Institute for Food and
Agricultural Sciences and should be
used only to trace the historic work of
the Institute and its staff. Current IFAS
research may be found on the
Electronic Data Information Source

site maintained by the Florida
Cooperative Extension Service.

Copyright 2005, Board of Trustees, University
of Florida

Bulletin 257 March, 1933

Wilmon Newell, Director


Director of the Laboratories, The New York Botanical Garden.

Bulletins will be sent free upon application to the


John J. Tigert, M.A., LL.D., President of the P. K. Yonge, Chairman, Pensacola
University A. H. Blanding, Bartow
Wilnon Newell, D.Sc., Director Raymer F. Maguire, Orlando
H. Harold Hume, M.S., Asst. Dir., Research Frank J. Wideman, West Palm Beach
Harold Mowry, B.S.A., Asst. Dir., Adm. Geo. H. Baldwin, Jacksonville
J. Francis Cooper, M.S.A., Editor J. T. Diamond, Secretary, Tallahassee
R. M. Fulghum, B.S.A., Assistant Editor
Ida Keeling Cresap, Librarian
Ruby Newhall, Administrative Manager BRANCH STATIONS
K. H. Graham, Business Manager
Rachel McQuarrie, Accountant NORTH FLORIDA STATION, QUINCY

L. O. Gratz, Ph.D., Plant Pathologist in Charge
R. R. Kincaid, M.S., Asst. Plant Pathologist
MAIN STATION, GAINESVILLE W. A. Carver, Ph.D., Associate Agronomist
R. M. Crown, B.S.A., Assistant Agronomist
AGRONOMY Jesse Reeves, Farm Superintendent
W. E. Stokes, M.S., Agronomist**CITRUS STATION
W. A. Leukel, Ph.D., Agronomist CITRUS STATION, LAKE ALFRED
G. E. Ritchey, M.S.A., Associate* John H. Jefferies, Superintendent
Fred H. Hull, M.S., Associate Geo. D. Ruehle, Ph.D., Associate Plant Pathol-
J. D. Warner, M.S., Associate ogist
John P. Camp, M.S., Assistant W. A. Kuntz, A.M., Associate Plant Pathologist
B. R. Fudge, Ph.D., Associate Chemist
ANIMAL HUSBANDRY W. L. Thompson, B.S., Assistant Entomologist
A. L. Shealy, D.V.M., Animal Husbandman*
R. B. Becker, Ph.D., Specialist in Dairy Hus- EVERGLADES STATION, BELLE GLADE
W. M. Neal, Ph.D., Associate in Animal Nutri- R. V. Allison, Ph.D., Soils Specialist in Charge
tion R. N. Lobdell, M.S., Entomologist
E. F. Thomas, D.V.M., Assistant Veterinarian F. D. Stevens, B.S., Sugarcane Agronomist
W. W. Henley, B.S.A., Assistant Animal Hus- G. R. Townsend, Ph.D., Asst. Plant Pathologist
bandman B. A. Bourne, M.S., Sugarcane Physiologist
P. T. Dix Arnold. B.S.A., Assistant in Dairy In- J. R. Neller, Ph.D., Biochemist
vestigations A. Daane, Ph.D., Agronomist
R. W. Kidder, B.S., Asst. Animal Husbandman
CHEMISTRY AND SOILS Ross E. Robertson, B.S., Assistant Chemist
R. W. Ruprecht, Ph.D., Chemist"*
C. E. Bell, Ph.D., Associate
J. M. Coleman, M.S., Assistant H. S. Wolfe, Ph.D., Horticulturist in Charge
H. W. Winsor, B.S.A., Assistant W. M. Fifield, M.S., Assistant Horticulturist
H. W. Jones, M.S., Assistant Stacy O. Hawkins, M.A., Assistant Plant
C. V. Noble, Ph.D., Agricultural Economist**
Bruce McKinley, A.B., B.S.A., Associate
M. A. Brooker, Ph.D., Associate FIELD STATIONS
Zach Savage, M.S.A., Assistant
ECONOMICS, HOME M. N. Walker, Ph.D., Plant Pathologist in
Ouida Davis Abbott, Ph.D., Specialist** Charge
L. W. Gaddum, Ph.D., Biochemist W. B. Shippy, Ph.D., Associate Plant Pathol-
C. F. Ahmann, Ph.D., Physiologist ogist
K. W. Loucks, M.S., Asst. Plant Pathologist
ENTOMOLOGY J. W. Wilson, Ph.D., Associate Entomologist
J. R. Watson, A.M., Entomologist** C. C. Goff, M.S., Assistant Entomologist
E. F. Grossman, M.A., Entomologist
A. N. Tissot, Ph.D., Associate Plant City
H. E. Bratley, M.S.A, Assistant A. N. Brooks. Ph.D., Plant Pathologist
P. W. Calhoun, Assistant, Cotton Insects R. E. Nolen, M.S.A., Asst. Plant Pathologist
A. F. Camp, Ph.D., Horticulturist"* A. S. Rhoads, Ph.D., Plant Pathologist
M. R. Ensign, M.S., Associate
A. L. Stahl, Ph.D., Associate Hastings
G. H. Blackmon, M.S.A., Pecan Culturist A H. Eddins, Ph.D., Asso. Plant Pathologist
C.. Van Cleef, M..A., Greenhouse Foreman A. H. Eddins, Ph.D., Asso. Plant Pathologist
C. B. Van Cleef, M.S.A., Greenhouse Foreman
W. B. Tisdale, Ph.D., Plant Pathologist** D. A. Sanders, D.V.M., Veterinarian
George F. Weber, Ph.D., Plant Pathologist Monticello
R. K. Voorhees, M.S., Assistant
Erdman West, M.S., Mycologist Fred W. Walker, Assistant Entomologist

*In cooperation with U.S.D.A. Bradenton
"**Head of Department. David G. Kelbert, Asst. Plant Pathologist



1. The rules of behavior for the A group .......................... 6

2. The rules of behavior for the B group ........................ 10

3. The daily reciprocations between A and B varieties ........... 10

THE LISTS OF A AND B VARIETIES................................... 12


THE NORMAL CYCLES OF DIANTHESIS ................................ 16


CHANCES FOR SELF- AND CROSS-POLLINATION ........................ 26



FRUIT SETTING BY TENTED TREES ................... .............. .. 32


ON INTERPLANTING AVOCADOS.................................... 33

1. Factors determining cross-pollination in avocado orchards..... 34

2. Other factors determining fruit production................... 36

3. Methods of interplanting avocados............................ 37

4. Some practical matters regarding interplanting............... 39

ACKNOWLEDGMENTS ................... ........................ 42

LITERATURE CITED ............... .............................. 44

Fig. 1.-Avocado flowers in the first opening, a, Pistil with stigma fresh
and receptive; 1, flower bud; 2, flower open in first period; 3, flower in
the interval between first and second opening. Enlarged about two


Pollination is the process by which pollen is carried from the
stamens and deposited on the stigmas of the pistils in the flowers
of plants. Then the growth of pollen tubes through the stigma
to the ovary leads to the final acts of fertilization, which include
the fusion of a sperm cell from the pollen tube with an egg cell
in the ovule. This double cell develops into the embryo of the
seed. It is certain that it is the rule that a fruit with a seed
can mature on most of the avocados of the present day only when
there has been a proper pollination that is followed by fertiliza-
tion. Thus proper pollination constitutes an important step in
the production of fruit.
But in avocados there is a most remarkable regulation in the
development of the flowers and of their stamens and pistils which
restricts and even prevents many self-pollinations,' many close
pollinations,- and also many cross-pollinations.3
This development and regulation is such that the flowers of one
group of seedlings and clonal varieties (the "A" group) function
as pistillates or females in the forenoon and as staminates or
males in the afternoon while those of another group (the "B"
group) function as males in the forenoon and as females in the
afternoon. Thus there is an adaptation that restricts self- and
close-pollinations and that provides for an alternation of the
reciprocal cross-pollinations possible each day between certain
members of the two groups.
But the horticultural practice of vegetative propagation by
grafting and budding to obtain clonal varieties of avocados that
are planted in solid blocks operates to prevent cross-pollination.
In nature, or in plantings of seedlings only, every avocado tree is
a seedling and any group of trees will be a mixture of A trees and
B trees with opportunity for reciprocal cross-pollinations. When
a seedling of merit is propagated by grafting or budding all trees
of the clonal variety are alike in their trunk and branches (except
for bud variation), they all have the same flower behavior, and
there is no more chance for pollination from tree to tree (intra-

'Pollen taken from the stamens to the pistil of the same flower.
2Pollen taken from one flower to the pistil of another flower on the same
plant or on another plant of the same clonal variety.
"Pollen taken from a flower of one plant to the flower of a different plant
either a seedling or a plant of a different clonal variety.

6 Florida Agricultural Experiment Station

clonal pollination) than there is for pollination from flower to
flower on a single tree. In such cases proper interplanting to
provide for natural cross-pollinations becomes desirable or neces-
How the types of flower behavior operate in the clonal varieties
now in cultivation, how they influence and determine the produc-
tion of fruit, the extent to which these adaptations for cross-
pollination make interplanting necessary or advisable, and what
particular interplanting will promote increased and maximum
yields of fruit in any or in all varieties are matters that are vital
in the culture of avocados.
The flowers of all avocados thus far studied, except the Collin-
son, have both pistils and normally developed stamens-that is
they are perfect flowers. Also the flowers are very similar in
structure and appearance. They are rather small and inconspicu-
ous in color; they are produced in large numbers; and a tree has
a succession of sets of flowers that gives continuous flowering for
at least several weeks and sometimes for several months. Each
flower has one pistil, two sets of stamens arranged in an inner
whorl of three and an outer whorl of six, and two sets of nectaries
corresponding in number and relative position to the sets of
The flowers open synchronously in sets that have, normally,
two distinct periods of opening. There is for each set (a) a period
of first-opening when the flowers function as females, (b) an
interval of being closed, and (c) a period of second-opening when
they function as males, and these three periods constitute what
may be called a cycle of dianthesis. The first-opening and the
second-opening take place on different days and they cover differ-
ent hours of the day. The succession of sets day after day brings
a daily alternation of a new set in its first-opening with an older
set in its second-opening and there is little or no overlap in the
periods of opening of the two sets.
Under favorable weather conditions the flower behavior for the
A group of seedlings and clonal varieties is as follows:
The First Period of Opening:-In the hours of the early morn-
ing no flowers are open anywhere on the tree but during the early

Bulletin 257, The Pollination of Avocados 7

forenoon flowers begin to open almost in unison here and there
in the various clusters of flower buds all over the tree. These
flowers have not been open before. In this first-opening of flowers
the six leaf-like segments of the perianth separate and bend out-
ward; the two sets of six outer and three inner stamens follow
and when the flower is fully open the stamens lie against the
perianth and are nearly at right angles to the main axis of the
pistil, as seen in the open flowers shown in Figure 1. Thus the
pistil stands erect, alone, conspicuous, and fully exposed with the
slightly expanded end (the stigma) white, fresh and evidently
ready to receive pollen. Soon nectar appears as a glistening film
or as droplets on the surface of the inner set of three nectaries
which stand erect from between the inner set of stamens. Now
bees and other insects seeking nectar can scarcely fail to brush
against the stigma. The pistil is ready for pollination but no
pollen is being shed from the stamens of the flower. The flower
is, for the time being, functioning only as a female.
Directing attention to the numerous flowers open on the entire
tree during the forenoon one finds that all the flowers which are
open are in the same condition as the flower just described. No
flowers are shedding pollen; all have pistils ready for pollination.
This condition continues throughout the forenoon during which
the entire tree functions as a female. About midday the numer-
ous flowers of the set that was open during the forenoon close
without shedding pollen. In doing this the perianth segments
fold inward over the pistil until the flower is completely and tightly
closed (see 3 in Fig. 2).
Thus an entire set of flowers, numbering several thousand for a
large tree in the maximum of blooming, has been open and in
the female stage during several hours of the forenoon, and during
midday these flowers have all closed almost in unison.
The Second Period of Opening:-During the hours of midday,
and usually while the flowers which were open in the forenoon
for their first period are in the process of closing, the flowers of
another set begin to open. To a casual observer it would doubt-
less appear that the same flowers open in the forenoon merely
continue open during the afternoon. But if a number of indi-
vidual flowers be tagged for identification the complete midday
shift of sets can readily be determined and also the two periods
of opening of a set of flowers can be observed and properly related
to the daily sequence or alternation.
The flowers of a set opening for their second opening in the

8 Florida Agricultural Experiment Station

afternoon are distinctly different in appearance from those open
during the forenoon (compare Figures 1 and 2). At this time
the end of the pistil is frequently dark-colored and sometimes it
is shrivelled and of course unable to receive pollen in proper polli-

Fig. 2.-Avocado flowers in the second opening, b, Pistil with stigmatic
end shriveled; c, stamens with uplifted valves on which pollen is ex-
posed; 3, flower in the interval; 4, flower open in second period; 5,
flower closed after the second opening. Enlarged about two times.

Bulletin 257, The Pollination of Avocados 9

nation. The stamens are now noticeably larger and somewhat
longer; the inner three stand erect in the middle of the flower
around and overtopping the pistil and facing away from it; the
outer set of six stamens stand at an angle of about 45 degrees.
Not long after these flowers open, pollen begins to be shed. The
pollen is ingeniously lifted out of each of the four chambers of an
anther by a spoon-shaped valve that opens quite like a trap door
and bends upward. A somewhat sticky mass of pollen is gently
held within the infolded margin of each valve, somewhat as one
might hold a ball of popcorn in an upraised hand. Thus the nine
rod-shaped stamens of each flower stand bristling in different
directions with pollen exposed in several directions from their
summits. (Fig. 2). Below at the base of the outer set of stamens
and between the stamens, a set of short-stemmed dome-shaped
nectaries excrete thick films of nectar. In their efforts to obtain
this nectar bees and other insects climb over the stamens, push
in between them, and their hairy bodies become more or less
smeared with the sticky pollen. But if pollen is not carried away
by insects the sticky substance about and between the grains
hardens and binds the pollen grains of each valve into a mass
which then soon falls to the ground.
A careful census of the many flowers open on a tree of the A
group during the afternoon will reveal that all the flowers that are
open are in the same condition. They all shed pollen with the
maximum of pollen shedding during the middle of the afternoon.
Late in the afternoon the flowers of this set close almost in unison
never to open again.
The Daily Alternation of Sets:-Thus during the hours of
daylight two different sets of flowers open and close on trees of
the A group and their periods of opening alternate. One set opens
for the first or female opening during the forenoon; another set
opens for the second or male opening in the afternoon. The ma-
turity of the pistil or pistils and the stamens of the same flower
at different times is known as dichogamy(14).* In avocados the
dichogamy is synchronous for the entire tree.
The Cycle of Dianthesis:-The A varieties have flowers which
open for the first period during the forenoon, close around midday
and remain closed during that afternoon, the night following and
the forenoon of the next day, and then open for the second period
during the afternoon of the next day. Thus for a single flower
*Numbers in parentheses (Italic) indicate references to literature to be
found on page 44.

10 Florida Agricultural Experiment Station

there are two distinct and separate periods of opening or anthesis,
a condition which may be designated as dianthesis. Between the
two periods of opening there is an interval of about 24 hours. The
entire time from the beginning of the first or female opening with
the end of the second or male opening (the completed cycle of
dianthesis) covers slightly less than 36 hours.
The succession of sets continues to operate on this schedule very
regularly under favorable weather conditions and this brings two
different sets into alternation or daily sequence each day. Each
forenoon a new set opens for the first or female opening and this
same set opens for the second or male opening on the afternoon
of the following day. Each afternoon the set which opens for
the second opening had its first period of opening during the fore-
noon of the previous day.
Each flower is perfect, it has two distinct periods of opening,
it is mature as a female during the first period and as a male during
the second period. There is the development of the two sexes in
each flower at different times (dcclogamy). The flowers open
synchronously in sets and the two periods of opening occur during
different hours of the day. Thus the flowers open on a tree at
any one time are alike, and the development of the flowers is so
coordinated or synchronized that the entire tree functions as a
female in the forenoon and as a male in the afternoon.
The flowers of varieties of the B group are like those of the A
group in structure and general appearance. Also they have the
same two periods of opening and the flowers also open synchro-
nously in sets. But the sets are normally open for the first or
female period in the afternoon and for the second or male period
in the forenoon either of the following day or of the second day.
The succession of sets of flowers operating in this cycle gives a
sequence in the daily alternation that is the reverse of that in
class A. The flowers of a tree of a member of the B group func-
tion as males in the forenoon and as females in the afternoon.
The relative flower behavior typical for A and B varieties and
the reciprocation in pollination which results may be shown by
comparing two clonal varieties such as Taylor and Panchoy.
During the forenoon, while flowers of Taylor are in the first
period or female anthesis, the flowers of Panchoy are all of the
second period or male opening (upper section of Fig. 3). Thus
trees of Panchoy are shedding pollen during the forenoon at the

Bulletin 257, The Pollination of Avocados 11

time when the flowers of trees of Taylor are most ready to be
pollinated. During the midday an alternation or change in sets
occurs for each and then in the afternoon the flowers that are
open on Panchoy are in the first period or female opening while

Fig. 3.-Flowers of Panchoy at left, of Taylor at right; upper for the fore-
noon, lower for the afternoon. 1, Bud unopened; 2, flowers in first
period; 3, in the interval; 4, in the second period; 5, after the second

12 Florida Agricultural Experiment Station

the flowers open on Taylor are all of the second period or male
opening (lower section of Fig. 3). These two varieties are, there-
fore, reciprocating in their daily alternations of sex. Taylor is
female in the forenoon while Panchoy is male and then while
Taylor is male in the afternoon Panchoy is female-all during the
hours of a single day.
The adaptations for reciprocal cross-pollination are marked and
evident. There is opportunity for pollen to be carried from Pan-
choy to Taylor during certain hours of the forenoon; during cer-
tain hours of the afternoon pollen can be carried from Taylor to
A total of 413 clonal varieties of avocados are listed by name
in the Year Book of the California Avocado Association for 1932.
The flower behavior for most of these has either not been deter-
mined or there has been no record made in publications. In the
following lists of A and B varieties several published lists have
been combined with additional data obtained in Florida in 1932.
Varieties which normally have flowers open for the first or
female opening in the forenoon and flowers open for the second or
male opening in the afternoon.
Atlixco Family Perfecto
Baker Flamingo Peterson
Barker Fuchs Pinelli
Baldwin Garcia Popenoe No. 3
Benik Gazzam Popenoe 51029
Blakeman Gottfried Puebla
Booth No. 1 Grande Reasoner
Booth No. 7 Hawaii Richardson
Brooks' Hawaiian Sharpless
Butler Hollis Simmonds
California Ishkal Sinaloa
Cantel Kanan Solano
Carlsbad Kashlan Spinks
Challenge Kay Taft
Lower Kosel No. 1 Taylor
Collason No. 1 Kosel No. 3 Taylorson
Collason No. 2 Kosel No. 4 Ultimate
Collason No. 3 Kosel No. 5 Wagner
Collinred C Kosel No. 6 Waldin
Collinred D Lula Ward
Collinred E Mahan Wester
Collinson No. 1 (pollen Manik S. P. I. 18729
sterile) Mayapan S. P. I. 19205
Collinson No. 2 McCann S. P. I. 26698
Colorado Moanaloa S. P. I. 26703
Dickey Mowry S. P. I. 29137
Dickinson Murrietta Green S. P. I. 29379
Dunedin Murrietta 2 lb. S. P. I. 44626
El President Pankay

Bulletin 257, The Pollination of Avocados 13

Varieties which normally have flowers open for the second or
male opening in the forenoon and flowers open for the first or
female opening in the afternoon.
Akbal Fulford Pollock
Big Tree (Seedling) Ganter Popenoe 51105
Bitte (Seedling) Hanson Queen
Booth No. 6 Hardee Rey
Booth No. 8 Harmon Rome (S. P. I. 34831)
Butternut Hosack San Sebastian
Cabnal Ishim Schmidt
Cardinal Itzamna Simpson
Champion Knight Steffani
Chisoy Kosel No. 2 Stephens Choice
Coban Kosel No. 7 Surprise
Colla Lamat Tertoh
Collinred A Linda Trapp
Collinred B Lyon Tumin
Collins Mattair Val de Flor
Collinson x Trapp McClure Verde
(S. P. I. 61740) McDonald Walker
Collinson x Trapp Meserve Whitcomb
(Broad leaf) Montezuma Winslow
Cook Nabal Winslowson (Rolfs)
Dorothea Nimlioh S. P. I. 2689
Eagle Rock Nimliohson S. P. I. 26700
Earle's Late Nirody S. P. I. Seedless 32400
El Oro Northrup S. P. I. 44856
Estelle Nutmeg
Fuerte Panchoy

The above lists include most if not all of the most important
varieties in culture at the present time in California and Florida.
It is of significance that the A and the B types of behavior are
apparently about equally distributed among avocado seedlings.


The relative flower behavior of numerous varieties, members
of the A group and members of the B group, may now be consid-
ered with reference to the exact time during the day when the
two sets open and close and the bearing which this has on the
reciprocations in pollination. In obtaining such data many flow-
ers are tagged with distinguishing numbers and observations are
made and recorded for trees of several varieties throughout the
same day. The observations are made from tree to tree as quickly
as possible, especially during the midday shift of sets, and the
condition of the flowers is indicated and checked on squared paper
for 15-minute intervals. The records for the day are then as-
sembled into a chart (see explanation of charts for symbols used)

14 Florida Agricultural Experiment Station

with the varieties arranged according to the time when the sets
of first-period flowers started to open. Such a chart gives a
graphic picture of what the flower behavior of the different vari-
eties has been during the day and what reciprocations for cross-
pollination are most complete.
The data for such charts have been collected for many days
of observation in Florida in 1925 and in 1932 and also in California
in 1923 covering the same varieties and often the same trees for
a series of successive days including normal behavior during
favorable weather and abnormal behavior of various degrees
during unfavorable weather. Figure 4 shows by diagrams the
normal flower behavior typical for A and B varieties.

March ri, ,95. Hours of the daY.
A. 7 o /l / 3 S ( 7 8
Par'-e c t-0
Aik -
Soo)a.O -
Taylor -- .. .
rai de --
-,iks. . .. ...-
"5pinks ---
Waqier -- . -
Ta --......
Hawaii -..
Wa Sd i -- -
13. ___
,lesev.ve .. . -.
Surprise --
Fanctiok -........-
WiMslo6w .
El Oro . -
LiUda- -
WH'nsowo .. .....
Fuert -
Pollock -
Tai'6olde i ---
Fig. 4.-A record of normal flower behavior in Florida for a day rather late
in the season of blooming. In this chart, as in those that follow, the con-
tinuous line indicates for each variety the entire time when flowers were
open for the first opening. The dots indicate the time during which flowers
of the second period of opening were shedding pollen and the dashes show
the time before and after the pollen was being shed when these flowers
were more or less open. As a rule pollen is shed most abundantly near the
middle of the period covered by the dots.
A survey of this chart clearly reveals the following important
features regarding the flower behavior of the several varieties

Bulletin 257, The Pollination of Avocados 15

1. There is a daily alternation in the synchronous dichogamy.
For each variety there were two different sets of flowers open
during the day. The flowers of one set functioned as females,
those of the other as males. Each set opened and closed in
unison and the two sets were open during different hours of the
day. While there was frequently a short interval of overlap of
the two sets during the shift in midday, the rule is that no pollen
was shed while flowers in the female condition were open for the
second period, then it was scant for a time and later the maximum
of pollen-shedding was reached after which pollen again became
increasingly scarce.
It may be noted that unless pollen is carried away by insects
the masses of pollen harden and become dried into little balls that
fall to the ground. It also seems most probable that the pistils
are not fully receptive for fertilization during the entire time
flowers are open for the first period and that in the charts a line
representing the most receptive condition of pistils would be much
shorter than the entire shown in the various charts. But con-
sidering the entire periods of opening, in every case the synchro-
nous alternation of sex as graphically shown in the chart most
decidedly limits self- and close-pollination.
2. There are two main groups of varieties. The varieties
studied on the date in question fell into two groups which recip-
rocated with respect to the relative sequence of the daily alter-
nation in the development of the two sexes. The members of one
group, called for convenience the A group, were female in the
forenoon and male in the afternoon; while the members of the
other group-B-were male in the forenoon and female in the
This record of daily behavior is typical and fully representative
for the various clons listed, for the A and B varieties in general,
and for the range of differences seen within the A and B groups as
these occur normally. Day after day during the more favorable
weather conditions the behavior of each of the varieties listed in
the chart will be quite similar.
3. There are varietal differences within the groups. It will
readily be noted in Fig. 4 that there are certain marked varietal
differences within each of the A and B groups in respect to the
exact time when the sets of flowers open and close. These are
features that are characteristic of the respective varieties. This
aspect of flower behavior was first noted and reported by

16 Florida Agricultural Experiment Station

For the A varieties studied on the day of this record there was
a difference of more than an hour in the time when the flowers
started to open for the first period. These flowers of Perfecto
opened at least 11/2 hours earlier than flowers on Waldin, and thus
there was some chance for the first-period flowers of the late
Waldin to be pollinated from the second-period flowers of earlier
varieties as Grande and Perfecto. In general, however, the hours
of opening were somewhat similar for all the A varieties, and
there was for all of these except Manik a short period of overlap
when the set in the first period was closing and the set in the
second period was opening.
For the B varieties very noticeable differences appear in the
time when the flowers open for the first period. Thus on this day,
for Meserve the first opening started early in the afternoon and
was completed before the first opening began on Taft's Golden.
Also for Trapp and Pollock the period of first opening came late
in the afternoon and continued until after dark. For El Oro and
Winslow the period of second opening was a relatively short in-
terval. For some of the B varieties there is a slight overlap of
sets, but for the ones listed at the bottom of the chart there is
actually an interval of several hours after the closing of the set
in the second period before the new set starts to open-a feature
of the dichogamy which was observed by Nirody(5). All of the
B varieties are, however, quite alike during the forenoon in respect
to the opening for the second period.
The differences shown in the chart in the relative flower be-
havior of various varieties are remarkably constant day after
day. The bearing which these types of behavior have on polli-
nation will be discussed later.
4. Reciprocations in pollination. Thus any variety of B fur-
nishes pollen at a time suitable for the pollination of any A variety.
But some of the B varieties have the period of first opening late
in the afternoon after pollen shedding has ceased or has almost
ceased by the A varieties, and for these there is a decided limita-
tion in the chances that ANY pollinations can be made during the
first period of opening. The relation which this habit of late
afternoon opening has to the second opening may now be viewed
in a consideration of the cycles of the sets of flowers.

The entire time from the first opening to the final closing of a
single flower or of a set of flowers, which includes the two periods

7 fours of]av |6 8 10121214 6 28 10112124 12 4 6 21 8 1 0122 1 4 6 8 10 4 8 110 1214 2 6
1 1 1 ljonC C CIC

c -cc cccccI + 1 { I
c erte : CcC ccc i- - o e
Theintervaln c c oC C Ca t

aele C CC C CCC --- -- -----
--c cc ccc 2cc .
-__-- i/or-- .. .. -----r.. c- ----C C C C c C C C C -I -,_ ---- --

-mrma- -- ._-_:_ i c"c clcc -1 ..- "

S .. 'i;c-;_ te__ .I I i ... I L I i_ .-:: ecI-c Ic - t -

-^ i ,r_ .. ...4Ii c ,--,c cc --I c) ."-i^ ;
--r -----^ 0^0!0!0!0"D^ IQ

Fig. 5.--Chart showing diagramically the main types of normal and abnormal cycles in the opening of the flowers of avo-
cados. An interval of time covering four periods of daylight and three intervening periods of darkness are represented .
The interval between the two openings of flowers is indicated by the letter C. Under abnormal opening the barred dash --
(+) indicates single opening.

18 Florida Agricultural Experiment Station

of opening and the interval of being closed, may be called the
cycle of dianthesis, the flowering cycle, or merely the cycle. In
considering these as represented in Fig. 5 one follows the behavior
of the flowers of a set continuously from the time they first open
until they close never to open again.
The shortest normal cycles of dianthesis are seen in the B
varieties that habitually open for the first period late in the
afternoon as do the Trapp and the Pollock but the B varieties
which open earlier in the afternoon, as Eagle Rock, have a cycle
that is slightly longer. The A varieties which regularly open for
the first period latest in the forenoon, as the Waldin, have a some-
what shorter cycle than those like Atlixco and Taylor that start
earlier in the day. The chart shows that there are more marked
differences between varieties in the hours of the first-period open-
ing than in the hours of the second-period opening. Thus when
A and B varieties are selected and arranged according to the time
of first-period opening, based on data collected on the same day
or on comparable days, there is, as shown in Fig. 5, a rather grad-
uated series from Trapp to Atlixco and the break between Eagle
Rock of the B group and Waldin of the A group is less than that
between the earliest and the latest within each of the groups.
But in respect to the hours of the second opening the A and the B
groups are more completely exclusive.

The particular behavior of sets of avocado flowers opening on
any one tree as to the precise time of opening, the duration of an
opening, the period of overlap and the lapse between sets, and
the length of the cycle is affected by weather conditions and par-
ticularly by changes in temperature. In extreme cases the entire
sequence of normal behavior is thrown entirely out of stride and
the action of the flowers becomes very irregular. Various sorts
and grades of such behavior are shown diagrammatically in Fig. 6
and in the charts which follow it.
Single Opening:-After several days of relatively high and
rather uniform range in temperatures a sudden lowering of tem-
perature may cause such short-cycle varieties as Trapp and Pollock
to omit the first or female opening of an entire set or of part of
a set and then this set of flowers may open the next day for a single
opening. Then the flowers may shed no pollen at all, they may
shed pollen poorly, or they may sometimes shed pollen freely and
quite fully, depending on the weather conditions.

Bulletin 257, The Pollination of Avocados 19

Skutch(9) has also observed that more pollen-shedding flowers
may be open on certain trees during a forenoon than there were
flowers in the first period during the preceding afternoon. This
condition has frequently been seen by the writer and the obser-
vation made that it may involve flowers which open but once for
a pollen-shedding period or it may involve flowers that operate
on a 48-hour cycle, as reported under "B2" and "B3" types of
flower cycles(13). Sometimes counts will show more flowers
open on B varieties during the afternoon than in the following
forenoon in which case certain flowers may not open for the
second opening or they may do so on the second day following.
That some varieties may be quite regularly A or B while others
of B are more erratic is a frequent observation (see Figs. 6 and 7).
This condition illustrates varietal differences.
Another type of single opening occurs when first-period flowers
do not entirely close during the night but enter into an extended
single opening. If the following day is favorable such flowers
may shed pollen.
The writer has carefully studied single-flowering and diligently
searched for seedlings or varieties which may have ONLY single
opening in the hope that this might be associated with self-fruit-
ing. This behavior is strongly developed in the "late" B varieties,
but no variety has been found which maintains single opening to
a noticeable degree under the most favorable conditions of
Delayed Opening and Reversal of the Daily Alternation:-
Under unfavorable weather conditions the first or the second
opening is often delayed; the first opening of A varieties may be
delayed until in the afternoon, and that for B varieties may be
delayed until the following forenoon and thus the normal daily
sequence is for the time being temporarily reversed (see espe-
cially Fig. 9). A set of flowers that would normally open during
any one day may be delayed until the following day or until the
second day. In extreme and continued inclement weather the
development and maturity of flowers is greatly retarded and the
opening of flowers in sets that are synchronous and alternating is
scarcely in evidence.
. Extended Cycles:-Unfavorable weather conditions or marked
changes in weather while a set of flowers is open often extend the
cycle of dianthesis either in the period of first opening, in the
interval, or in the second opening, or in two or more of these
conditions. Various types of this are shown in the table of

20 Florida Agricultural Experiment Station

Fig. 5. For example, a single set of flowers of Atlixco which
normally has a cycle of about 36 hours may have an extended
cycle of 80 hours.
Under certain conditions some of the B varieties, as the Fuerte,
will have the sets of flowers operating in a cycle of 48 hours, but
with a daily periodicity and alternation that appears to be quite
normal. This appeared to be the rule of the behavior for certain
B varieties in California during February, March and April of
1923 in those orchards where the writer made continued observa-
tions. Again this same variety and even the same tree that is
operating on a 48-hour cycle will shift to a short cycle of 24 hours.
One of the most surprising types of behavior is that in which
part of the flowers of a set have a cycle of 24 hours and the rest
have a cycle of 48 hours. In this case nearly always the pistils
of those flowers opening for the second opening in a cycle of 48
hours are at that time blackened and obviously unable to function
to pollination and these flowers open as much as an hour in advance
of the flowers of the shorter cycle. The presence of pollen-shed-
ding flowers of both short and long cycles may make decided dif-
ferences in the number of flowers counted on successive days.
In the present discussion it is considered that the 48-hour cycle
for B varieties may be classed as abnormal but it should be noted
that the flowers of a variety may continue in this cycle for some
time. I
The normal and the abnormal behavior of numerous varieties
have been studied repeatedly in the same grove on different days.
Typical records for such studies are presented in Figs. 4, 6, 7 and
8 and it may be stated that the observations recorded in these
charts were all taken in the same grove (that of the late Wm. J.
Krome at Homestead, Florida) and for the same tree when any
one variety is noted in different charts.
The record of Fig. 6 is for a day following rather low night
temperatures. There was decided delay in the opening of both
first-period and second-period flowers and for several of the B
varieties which normally have first-period flowers opening late
in the afternoon there was omission of this set and continued
opening (indicated by arrow heads) from the previous afternoon.
But for most varieties there was the daily alternation of two
sets complete, there was no overlap of sets when pollen was being
shed and the reciprocation between many A and many B varieties
was good. On this day there were no first-period flowers open

Bulletin 257, The Pollination of Avocados 21

Mayady to, if.I. Hours oS the dai .
Class A. 7 7f /o / 1 / 3 Y { 7 y
Perjecto -
Spi ns
"Taylor --
/Manik --
Dickinso --
Wagni er
Qrav de.
Ta t
Wa Idi rn
Class B.
Eagle ?ock. -
Mesee - -
Suvprise --
ScloLJ -. .....-
Sc idt -----

Verde -
Nimrlioh -
Winslowso .. -
Hardee -
FuerTe >----. ) -->
Pollock -
Trapp .
Este Ile - -- -..
Tact's Goldev>- - - -
Fig. 6.-Record of flower behavior for a day after a night of somewhat
unfavorable temperatures.

on Estelle and Taft's Golden. Some of this set opened for a single
opening during the next day.
Figure 7 is for two successive days and illustrates the effects
on flower behavior of a sudden drop in temperature to a maximum
of 770 and a minimum of 460 after several days of temperatures
ranging about 100 higher. The opening of sets was delayed but
the A varieties and most of the B varieties had two sets in daily
alternation. The late B varieties were most irregular with single
opening of both types and with much extended opening. On both
days Estelle had only flowers of a single opening, and Trapp had
most of its flowers in a single opening. On the day of this record
there was for many varieties no overlap of sets and for none was

22 Florida Agricultural Experiment Station

Mhavch f4, 25as. Hours of the daq.
A 7 ? 7 /0 // /2 2 5 6 7
Prfecto -- * --
Solao * * -
Crande ------ * -
Pinelli -....--- -

Waper .. -

Panchoy - -. ..
Aleserve -* -
Surprise .....--
Schmidt .. .-
uee -
Hardee -+- -+- -. -.-
Fuerte -
Wins lowsonl+- -t _--
Verlde -- -
Harwro >- -
Pollock + +- -- --
Estella --- -+- *
"Tra pp -- -- -

ieaJIeI -+- .+. --P
extended opening, aGond sindegle opening.M
/ alrch .5s, / 9zs
A 7 6 y /o //2 3 5- 6 7
Dickinson -
Atixco .
WaTd in . . .

aeser ve- *
Surprise --- * ---
Winslowsoni -- --
Fu e rte --
Hardee -- --
Tr>pock ... -
Es tWle -- -- - -

Fig. 7.-Records of flower behavior for two days showing delayed opening,
extended opening, and single opening.

Bulletin 257, The Pollination of Avocados 23

there overlap for a considerable portion of the period of pollen
Various sorts of extreme off-stride flower behavior, frequently
observed both in California and in Florida, are shown graphically
in Fig. 8. The 9th and the 10th of February had been warm with
Feb. 1., f25'. Hours of The. day.
A 7 6 /0o /21 12 3 f 7 ?
Affixco -
Solan o --

Fuevti ie r
Wag ei

Harmon -- ->
Winslowson -
North vp .. ->
Trap -- -
Feb. 13, lf2s.
A 78 9 I/ 1 / 20. 3 1 5 7 8
Atlixco --
Solano -- -
W.uer if ne --- ---
Finerte .
Haimon 4- >-
Winslow son _-- -- -__
Surprise -- -
Nor'hrup V -
Hatdee -- -
Pollock --
Trapp --
Fig. 8.-Record for two successive days showing much delayed and extended
opening of flowers.

maximum temperatures of 84 and 910 and the flower behavior
had been normal and regular quite as shown in Fig. 10. Heavy
rain on the llth was followed by clear cold weather and northerly
winds with a drop to a maximum temperature of 64 on the 12th,
and on the night preceding the 13th there was the only frost
experienced during February and March of that year in the
orchard where the observations were made. As the chart indi-
cates the two sets of flowers, which in favorable weather would
have opened and closed on the 12th, actually required two days
for their opening. There were both delayed and extended an-

24 Florida Agricultural Experiment Station

theses and most of the time the flowers were semi-open with feeble
and incomplete action of the flower parts and the shedding of
pollen was slow, irregular, and incomplete. On the 12th there
was no overlap of sets with one shedding pollen for any variety
studied. On the 13th there was decided overlap for Atlixco, Wag-
ner, Fuerte, Harmon and Pollock.
Thus the daily behavior of any one tree may differ from day
to day in response to weather conditions. With continued inclem-
ent weather the flowering may be erratic for several days, for
several weeks, or even throughout most of the entire season of
blooming. Then it may be difficult to determine whether varieties
are to be classed as A or B. From certain of the records obtained
in California the behavior of the same trees of three A and three
HouYs of the dauY
8' 9 /o // /" 3 1 S 6
Afay S. If93.

13 Fncho q...
SLinda -- -

Mat 7,i, ifz3.
'Shiarpless -. -
A icknai -- - --
lTat --
(Queen ----- -
13 IRFIviclIou -
Li nda- -- -

MaL, 'f, /123.
Shiaipless --
A icso - ---
"[Tat- -- -
_ (Queet ---
i Panchoq- --
Fig. 9.-Records of flower behavior for six trees on three different days.

B varieties on three different days have been combined in Fig. 9
to illustrate the point in question. On May 5 the daily alternation
of dichogamy was complete and in normal sequence. On May 17
under unfavorable weather there was for Sharpless and Taft a

Bulletin 257, The Pollination of Avocados 25

reversal of the daily sequence, no pollen shedding by flowers of
Dickinson, extended opening and overlap for two sets on Queen
and Panchoy, and for Linda only one set was open. During the
forenoon of the 19th, there was the return of sunny weather and
warmer temperatures. For the A varieties the set of flowers
delayed or continued from the previous day shed pollen and closed
about noon, then first-period flowers were open and these were
followed by a set of second-period flowers. Of the B varieties
Queen had three different sets of flowers open but Panchoy and
Linda had but one set and these were second-period flowers.
The various conditions of abnormal flower behavior were fully
described and discussed and some of them illustrated by photo-
graphs and charts in the first reports made by the writer(10, 11,
12) as well as in a more recent publication(13). Others have re-
ported and discussed such behavior (see especially 7). Erratic
behavior, often to the degree that there was no definite daily alter-
nation, was the rule in California during February, March and
part of April in 1923, wherever the writer made observations.
Flower Behavior Throughout the Season of Flowering:-Dur-
ing 1925 the flower behavior of a tree of the Taft variety and a
tree of the Trapp variety was observed carefully for most days
of the season of bloom.
For the tree of Taft (see Fig. 10) there was considerable varia-
tion in the hour when the sets of flowers opened but there was
/fours of the- da. .
7 s f /o // /2 / 2 3 5 6( 7 9 9

'Ir3b.257 __ __ * -
Mar. 2

'2 - ----- --
/0 ....

,, /I

"3,1 -------------- -

",1 / -- -

nl ---- . -
" % -------- .....
Fig. 10.-Records of the flower behavior for trees of the Taft variety in one
locality in Florida during the season of 1925.

26 Florida Agricultural Experiment Station

for each day a definite alternation of two sets. Except for a few
minutes on the first day of record there was no opportunity for
pollen to be carried from second-period flowers to first-period
flowers open on the tree. The flower behavior was most normal
during the warmer weather of the last half of the season of bloom
and then the shift of sets occurred more nearly at the noon hour.
The record for the tree of the Trapp variety is shown in Fig. 11.
Hours of tie day.
7 b f /0 / / a 3 1Y 5 7 r f /o
Feb./ -- -
",1 -- "
",, /f .

25 --

-- -- -
,, 5- ._L --.

/ .

2 / .


7 - -- --

Fig. 11.-Records of the flower behavior for trees of the Trapp variety in one
orchard during the season of 1925.

There were three days (Feb. 11 and 13 and March 3) when a set
of first-period flowers failed to open. Only on two days did the
first-period set open as early as 3:00 p. m.; usually this set opened
late and remained open until after dark. At no time throughout
the many days of observation were there any first-period and
second-period flowers open together.


Self-Pollination:-In avocados self-pollination of individual
flowers can occur only during the normal second-period of dian-
thesis or during a single opening when pollen is being shed. Then

Bulletin 257, The Pollination of Avocados 27

considerable pollen of both self- and close-pollinations must, it
would seem, reach the stigmas especially as a result of the activi-
ties of insects. If the stigmas are still receptive, the conditions
favorable for pollen-tube growth, and the ovules and their eggs
still ready for fertilization and responsive to self fertilization
some fruit should result.
It would seem that the B varieties with shortest cycles, such as
Trapp and Pollock, would be most likely to set fruit to pollinations
made during the second period of the cycle. The flowers open for
the first period late in the afternoon when pollen is not abundant
on A varieties and the chances for any kind of pollination at this
time are greatly reduced. If the pistils remain receptive until
the next forenoon self-pollination or cross-pollination with other
B varieties may be responsible for some or even all of the fruit
that may be set at certain times by such varieties.
It would seem that the flowers of the A varieties are less liable
to remain receptive to pollination until the second opening. In
fact the stigmas of such flowers are often disintegrated, black-
ened, and shrivelled and obviously in no condition to function to
any kind of pollination. Such flowers, it would seem, must be
properly pollinated during their first opening if they are to yield
fruits. The same is true of the B varieties for those flowers that
open in a long cycle of 48 hours.
Close-pollinations (from flower to flower on same tree) may be
of two sorts; (a) between different flowers all of which are shed-
ding pollen, or (b) pollination of first-period flowers from pollen
of flowers in the second-period. The latter may be designated as
dichogamous close-pollination.
The first mentioned type can occur at all times when flowers
are shedding pollen and is presumably quite the same as self-
pollinations except when a short-cycle set and a long-cycle set
are both shedding pollen together. It is certain that for many
varieties these pollinations do not lead to the setting of fruit.
Dichogamous close-pollinations can occur (1) during an over-
lap of sets and (2) when insects force flowers during the interval
when they are closed after the first opening. During normal
flower behavior there is, for most varieties thus far studied, very
little overlap of sets during the time when pollen is shed. For a
large majority of the flowers opening normally there is no chance
for close-pollination of first-period flowers. The rule of normal
behavior is that while the two sets may overlap slightly there is
no pollen shed by flowers of the second opening during the time

28 Florida Agricultural Experiment Station

of the overlap. This is amply shown in Figs. 4, 6 and 7 for the
daily behavior of numerous varieties and also in Figs. 10 and 11
which record the flower behavior of individual trees of the Taft
and the Trapp varieties during a season of bloom.
During various types of irregular and erratic flower action
there may be many chances for dichogamous close-pollination.


Pollination is but a first step in the series of events that are
necessary in most avocados for fruit production. It is followed
by the processes of fertilization including the growth of pollen
tubes in the pistils and the fusion of sperm and egg in the ovule,
all of which are processes which involve intricate physiological
In dichogamy, such as occurs in avocados, the condition of the
pistil, and especially of the stigma at the time when pollen reaches
it, is a factor which determines whether pollen tubes will grow.
It seems clear that the rule is that in avocados the pistils are most
receptive to pollination during the first period of a normal flower
behavior, and that many pistils are less able or entirely unable to
set fruits to any pollinations during the second period of opening.
In many plants, of which the apples, the pears and the sweet
cherries may be cited, it is known that self-pollinations and also
certain cross-pollinations made at the time when pistils are fully
receptive are not effective in fertilization and in such cases only
pollinations between varieties that are compatible in fertilization
result in satisfactory sets of fruits. Certain varieties may be
somewhat self-fruitful but will yield much larger crops to proper
cross-pollination. Whether somewhat similar conditions exist in
avocados in addition to the dichogamy is not definitely known.
At the present time little direct knowledge is available on the
processes and events of fertilization in avocados. H. van Elden
found (reported by Hodgson, 4) that for various Mexican and
Guatemalan varieties the egg apparatus is developed and ready
for fertilization when flowers are in the first period of opening
and he found cases of fertilization at this stage. This is what one
would expect from the flower behavior, but it does not necessarily
indicate that the pistils of a few flowers on many varieties or that
the pistils of many flowers of certain varieties do not remain more
or less receptive until the second opening.
During 1932, T. W. Young in cooperation with the writer made

Bulletin 257, The Pollination of Avocados 29

studies of several varieties as to the structure of the pistil and
the growth of pollen tubes in the style after various sorts of polli-
nation. These studies indicate that there may be abnormalities
in the structure of certain pistils which reduce or prevent their
functioning in fruit production. It is possible that further study
will show that these have a relation to abnormal flowering, to con-
ditions of climate or culture, or to season of flowering or that they
are rather characteristic of certain varieties. It also appears that
the number of pollen tubes which can penetrate through the style
is restricted to few tubes or possibly to one tube. Further studies
of the actual processes of fertilization should be made in respect
to determining the conditions essential to the production of fruit,
and to learning just what pollinations are most likely to yield
There is a tendency for a few varieties, as the Fuerte, to pro-
duce fruits which contain no seeds. Two clons which are said to
produce only seedless fruits (F. P. I. Nos. 14889, and 14890) were
obtained from Mexico in 1905 for trial in the United States. An-
other avocado, F. P. I. 32400, from C. P. Taft of Orange, Califor-
nia, is described as having "fruit absolutely seedless but very
small, 3 to 5 inches in length and 3/ to 1 inch in diameter." A
tree of the F. P. I. 32400 is now growing in the Plant Introduction
Garden at Chapman Field, but I am informed by T. A. Fennell in
charge of the Garden that it has not yet bloomed. It is not
definitely known whether such fruits develop without pollination
as parthenocarpic fruits (as do the fruits of the naval orange) or
whether there is death of embryos after fertilization as is the
case for certain seedless grapes. The seedless fruits which appear
on avocados are relatively small and somewhat irregular in shape
and give little promise of commercial value. It is possible, how-
ever, that good seedless sorts whose fruits are parthenocarpic
may in time appear or be developed by selective breeding.
It has several times been suggested to the writer that certain
avocados may be able to produce fruits containing seeds without
any pollination or fertilization. Such a condition, which is called
apogamy, is known to occur in relatively few of the seed plants.
There is at present no proof or even very suggestive evidence that
apogamy does occur in avocados. That certain trees of the same
clon, as the Fuerte, are highly self-fruitful through apogamy in
one orchard while other trees a short distance away are not is
very improbable and is not to be accepted unless proven.
In summary, it maybe emphasized that in the chain of events

30 Florida Agricultural Experiment Station

which lead from flower to fruit in avocados there are several weak
links due in large part to the element of chance in the play of
circumstances. Chances for fruit development depend on the
presence of a seed. Chances for a seed depend on the presence of
an embryo in the very young fruit. Chances for this embryo
depend on proper fertilizations. Chances for fertilization depend
on proper pollination which in turn depends on flower behavior,
on the reciprocation of varieties in pollination and on the activities
of insects. On all links of this chain the external conditions of
weather and cultural conditions and the internal conditions of
nutrition and development exert their various influences. Proper
pollination, fertilization after pollination, and the holding and
maturing of fruit are three weak links in this chain.


Self- and Close-Pollinations:-In California during 1923 several
thousand self- and close-pollinations were made by hand mainly
for the varieties Harmon, Fuerte, Northrup, Puebla, Dickinson and
Taft, but for these not a single fruit was set. For Carton and
Queen small fruits set, even when flower clusters were enclosed
in paper bags and when there was no pollination except possibly
that of incidental selfing, but none of these matured.
Cross-Pollinations:-As previously reported (Stout,12 and 13)
it is to be recognized that relatively few cross-pollinations which
are made by hand result in fruit. Of a total of 3,430 carefully
executed cross-pollinations made in California during February,
March, April, and May, 1923, only 173 flowers started to set fruit
and some of these failed to develop. Similar results were obtained
for pollinations made in Florida in 1925. In 1932 T. W. Young
and the writer made 63 different kinds of controlled pollinations
involving a total of 1,151 flowers. Of these flowers 754 were cut
for microscopic study of the processes of fertilization. Of those
left on the tree, 45 selfs promptly fell without even starting to
form fruits and of the 352 flowers cross-pollinated only 14 set
fruits. None of these 14 matured, but had all fallen a few weeks
later (last check made 4-20-32). The cross-pollinations were
reciprocal and they were nearly all made during normal flower
behavior. It was later noted that considerable fruit set during
the same period to open pollination especially on Winslowson.
This fruit set rather abundantly at a time when its flowers were
opening normally, when there was no overlap of sets, when bees

Bulletin 257, The Pollination of Avocados 31

were frequent visitors and during which there was opportunity
for reciprocal cross-pollination with nearby and adjacent trees.
Evidently insect pollination was much more effective than hand
pollination. In certain cross-pollinations involving Lula, Wins-
lowson, Fuerte and Hawaiian the stigmas were stippled with a
fine stiff-haired brush to break the stigmatic glands with the
thought that such injury may be accomplished by insects and
may promote pollen-tube growth. But such pistils showed no
better pollen-tube growth than did unstippled pistils and the
flowers thus treated and left on the tree all fell. Such results do
not enable one readily to determine what are the relations of polli-

Fig. 12.-Showing fruit set by tree of Winslowson to open pollination during
a period when flower behavior was decidedly regular, when bee visitors
were frequent, and when reciprocating cross-pollinations were possible
from nearby trees. During the same period, careful self- and cross-polli-
nations on the same tree all failed to give sets of fruit.

32 Florida Agricultural Experiment Station

nation to the setting of fruit, what kind of pollinations are respon-
sible for the fruits which mature on a tree and to what erratic
setting of fruit is due.
Observation clearly shows that a large majority of the flowers
produced by avocados do not set fruits and this condition con-
tinues even after the careful cross-pollination by hand of first-
period flowers. Possibly (1) the pollinations are not made in a
way necessary for success, or (2) the pistils of many flowers are
unable to function in any relation except when conditions are
favorable to a normal development either in the structure of the
pistils or in their physiological condition, or (3) the most effective
relations in cross-fertilizations were not involved in the crosses
The extent to which avocado trees are able to produce fruit
without cross-pollination may, it would seem, be determined
experimentally by enclosing a tree in a cheesecloth "house" or
"tent" during the entire period of flowering. Also trees or parts
of trees of two varieties may be enclosed together. Bees to make
the pollinations may or may not be supplied.
Tests of this sort were first made for avocados at the Theosoph-
ical Homestead, Point Loma, California. The results have been
published and discussed(1, 2 and 3). In these tests Fuerte, Tins-
ley, and to some degree Dickinson were found to be self-fruitful
provided bees were enclosed to effect self- or close-pollinations and
it was concluded that at least the Fuerte is adequately self-fruit-
ful. There were somewhat conflicting results. For example,
part of a tree of Fuerte set no fruit when enclosed with part of
a tree of Spinks with two hives of bees. If trees of Fuerte are
adequately self-fruitful the enclosed portion of the tree in question
should have borne fruit to selfing irrespective of any cross-polli-
nation with the Spinks. The result seems to indicate either that
the insects did not effect cross-pollinations or that such pollina-
tions did not lead to fruit setting.
During 1925 in the orchards of Wm. J. Krome at Homestead,
Florida, a tree was tented with a hive of bees enclosed of each of
the varieties Linda, Panchoy, Taft, and Trapp(13) (Fig. 2).
About eight weeks after the close of the flowering period the Taft
tree had seven fruits, the Panchoy two, the Linda 22, and the
Trapp 18. There was no question that the flowers of these tented
trees received abundant visitations daily by the enclosed bees

Bull efin 257, The Pollination of Avocados 33

which effected self- and close-pollinations (including forced polli-
nation of first-period flowers) far in excess of what flowers of any
orchard tree is likely ever to receive. If these varieties readily
set fruit without cross-pollination the tented trees should have
set fruit more abundantly.
The flower behavior of adjacent trees of Taft and Trapp was
studied during the season (see Figs. 10 and 11). For each there
was a rather complete daily alternation with chances for self-
pollination or for close-pollination of first-period flowers only
through the forcing of flowers by the bees. The results obtained
with the tented trees show that almost no fruit was set to enforced
self- and close-pollinations and that proper cross-pollinations were
necessary for adequate fruit production.


The fruiting of the Collinson avocado is of special significance
in regard to the amount of cross-pollination which may occur
under orchard conditions. This clonal variety is completely pollen-
sterile(15, 18 and 13) and it furnishes no pollen for any sort of
pollination. Tests under tents(6) show that trees of Collinson are
unable to produce fruits without cross-pollination. This variety
has been rather widely planted in Florida where it is usually
highly fruitful. In fact, it received first place by vote in a ques-
tionnaire sent out to growers in 1931 as being very uniformly
fruitful and highly desirable for general culture. Robinson and
Savage(8, page 8) state that "The fact that this excellent variety
fruits well in mixed plantings is convincing evidence that cross-
pollination is the regular method of fruiting with avocados gen-
erally, and in this case it is the only possible explanation of fruit-
The fruiting of the Collinson may be confidently viewed as the
result of cross-pollinations with some other variety or varieties
and as evidence that when conditions are favorable cross-pollina-
tion may rather readily be accomplished.

The advantage to be gained by interplanting avocados lies in
the chance that there will be cross-pollinations which increase the
yields of fruit over that obtained in solid-block plantings of one

34 Florida Agricultural Experiment Stai.on

Three conditions are important in determining whether an
interplanting will be liable to increase the production of fruit.
1. The varieties interplanted should so reciprocate in flower
behavior that there is opportunity for cross-pollination.
2. Means for effecting the pollinations must be operating year
after year.
3. The pollinations when made must lead to fertilization and
to the development of fruit.
1. Any interplanting of an A with a B variety will increase the
chances for cross-pollination and these will be greatest when the
A and B varieties used most fully reciprocate in the hours of the
daily sequence and also bloom together over a considerable part
of the flowering season. On the basis of flower behavior the proper
interplanting of any variety will increase the chances for the
pollination of flowers especially when they are in the first-period
of anthesis and evidently most receptive to pollen.
An interplanting of only A or of only B varieties will give very
little increased opportunity over solid-block planting for the
normal pollination of first-period flowers. It is, however, quite
possible that certain short cycle B varieties as the Pollock and
Trapp will benefit by cross-pollinations with other B varieties dur-
ing the second-period of anthesis provided the pistils remain re-
2. An interplanting can only afford an opportunity for cross-
pollinations. Means for making the pollinations must be operat-
ing. Fortunately honey bees are very fond of the nectar of
avocado flowers and they also collect pollen from the flowers. They
freely visit the flowers during both of the periods of opening.
Various other insects visit the flowers of avocados and some of.
these may be important agents in effecting pollination. But hives
of the honey bee may be placed in the avocado orchards and to a
considerable degree kept under control and for this reason the
honey bee is to be considered as the most promising agent for
orchard pollination.
Insects in their visits to the flowers automatically effect many
self- and close-pollinations during the second period of opening
but it seems obvious-that the pistils of many such flowers are not
receptive. Theymay alsoect close-pollinations of first-period
flowers whenthese are possible through overlap or through forc-
ing. But to effect reciprocal cross-pollinations an individual

Bulletin 257, The Pollination of Avocados 35

insect must fly from one tree whose flowers are shedding pollen
to another tree whose flowers are in the first period of opening.
In general the insect must fly from B to A trees in the forenoon
and from A to B trees in the afternoon (see Figs. 3, 4, 6 and 7).
During 1932 in Florida a special effort was made to study the
activities of honey bees in their visits to avocado flowers. Mrs.
W. J. Krome and W. K. Walton obtained several hives for distri-
bution in the groves and E. Morton Miller and various of the
students of the Department of Zoology at Miami University made
special studies of the activities of the bees. One important fea-
ture which they determined is that certain bees collect pollen
only from avocado flowers while certain other bees from the
same hive collect nectar only. It is certain that many individual
bees that are active among the avocado flowers do not make cross-
visitations. They are liable to work on a single tree. Also it
should be recognized that those bees which collect only pollen
confine their attention to trees in the second period of the daily
alternation or to second-period flowers where there is overlap and
in so doing they make few or no reciprocal cross-pollinations. The
desired reciprocal cross-pollinations are made chiefly or only by
those bees which collect nectar and which in doing this fly directly
and repeatedly from a tree whose flowers are shedding pollen to
a tree of another variety whose flowers are in the first period.
Furthermore, it is possible that the qualities of the nectar pro-
duced by different varieties influence the activities of insects.
Nectar differs in color and apparently also in thickness and in
amount. The differences in the flavor and the odor of the nectar
produced by certain pairs of varieties may be such that individual
bees may not collect from the two during any one period of work.
Thus many bees, and other insects as well, may be very busy in
collecting nectar or pollen and yet be effecting no reciprocal cross-
It can readily be observed that honey bees do travel through
avocado groves to a considerable distance from the hive but
whether this is due to preference for some one variety or to some
other influence is not known.
Possibly bees are more liable to make cross-visitations at times
when the open flowers are relatively few, when bees are relatively
numerous, and when other pasturage is scarce. At certain times
bees may be attracted away from avocado flowers to other flower-
ing plants, especially to nearby citrus trees or even to cover crops.
Certainly avocados should not be inter-mixed with other fruit

36 Florida Agricultural Experiment Station

crops such as citrus which not only separate the avocado trees
but which may also be in flower at the same time.
In respect to the supply of hives of bees to be recommended for
properly interplanted orchards of apples, pears, or sweet cherries,
it is frequently considered that one strong colony to the acre is
sufficient (see Bulletin No. 577, New York State Agri. Experi-
ment Station). But in avocados the daily synchronous alternation
of dichogamy restricts opportunity for proper pollination and, it
would seem, makes it desirable to employ more than one hive to
an acre, at least for the present time until experience gives more
specific guidance.
3. The matter of affinities between varieties in respect to the
processes of fertilization after cross-pollination may be an impor-
tant one in the setting of fruit by certain avocados. The late
William J. Krome frequently stated to the writer that he believed
there is a decided cross-affinity between the Guatemalan and the
West Indian types. If this be true, certain pairs of varieties may
be more fruitful in reciprocal cross-pollination than are others.
This matter requires careful analysis by direct studies of fertili-
zation, by special tests with tented trees, and by comparison of
fruit yields in adequately interplanted orchards in which the
same varieties are employed in different combinations.
It is to be emphasized that reciprocating interplantings, ade-
quate pollination, and affinities in fertilization can not and will
not eliminate various other factors for unfruitfulness such as
habits of alternate bearing, unfavorable conditions of weather,
the attacks of fungi and of insect pests, and faulty methods of
culture particularly in irrigation, soil fertilization and pruning.
Such factors operate to influence yields of fruit and often they
are the limiting factors in fruit production. They are vital in
the culture of fruits like the navel orange which requires no
pollination whatever. In avocados they influence fruit production
in addition to the factor of proper pollination. Possibly in certain
localities unfavorable conditions of climate, of soil, or of culture
are the chief factors which are responsible for lack of fruit on
avocados and under certain conditions even the most proper recip-
rocal cross-pollinations may not be effective.
Frequently varieties of avocados set large numbers of fruits
(see Fig. 11) which soon fall. Some of this abscission of fruit is
due to physiological conditions either unfavorable to the retention
of the fruit or of the nature of competition between fruits for

Bulletin 257, The Pollination of Avocados 37

food and water. Some of the fruits which are without embryos
or which have dead embryos may be due to lack of pollination or
to abortion after fertilization and these are doomed to fall unless
there is development as seedless fruits. When there has been
proper setting of fruit the holding of fruit until maturity is no
doubt chiefly a matter of the influence of environmental and cul-
tural conditions.
The simplest type of interplanting is that which includes one
A variety and one B variety. In such plantings, three arrange-
ments of the trees may be considered as follows:

(2a) (2b) (2c)

In 2a and 2b a vertical row is composed of only one variety and
in 2c the trees of each variety are in diagonal rows. In plan 2b
the two nearest trees are always of the same variety but there
are four of the reciprocating variety next in the line of the diag-
onals. In 2a each tree has two adjacent reciprocations in the
same lateral row. In the plan 2c each tree has for its four
nearest neighbors trees of the other variety, and this would
afford opportunity for the greatest number of reciprocations in
cross-pollination. In any such interplanting the two varieties
should be in flower together at least during a considerable part
of the season of bloom and neither should be decidedly alternate
in bearing.
An interplanting of only two varieties gives opportunity for
only one kind of cross-pollination. The nectar of the two varieties
concerned may be so different that cross-visits by insects may be
reduced in number or it is possible that there may be lack of
affinity in the processes of fertilization.
When four varieties, two of the A group and two of the B group,
are interplanted the arrangement that seems best is as follows:
Al B1 A2 B2 Al B1
B2 Al B1 A2 B2 Al
A2 B2 Al B1 A2 B2
B1 A2 B2 Al B1 A2

38 Florida Agricultural Experiment Station

In this case each tree has immediately surrounding it in the four
nearest distances two trees of one reciprocating variety and two
trees of another reciprocating variety. Thus there are for each
tree chances for two different reciprocating cross-pollinations and
each operates from two trees. There is the chance that one of
the two cross-relations may be more effective than the other, and
hence the interplanting of four varieties is in general more likely
to give fruit than an interplanting of only two varieties. If,
however, cross-pollinations are not effective between any two, as
let us say Al and B1, then the Al and the B2 and the B1 ahd the
A2 should reciprocate. If it is determined that they do this per-
fectly the two pairs may be planted to greater advantage in a
two-variety interplanting.
When three A varieties and three B varieties are interplanted
the maximum opportunities for cross-pollinations are obtained
when the varieties of each group are arranged in repeating se-
quence in diagonal rows which alternate as shown in the following
diagram: Al B2 A3 B1 A2 B3
B1 A2 B3 Al B2 A3
Al B2 A3 B1 A2 B3
B1 A2 B3 Al B2 A3
Al B2 A3 B1 A2 B3
Bl A2 B3 Al B2 A3
In this arrangement any tree has as a female the chances for
three different inter-varietal reciprocations in pollination from
the four nearest trees and one of these is duplicated in the two
trees adjacent in the same vertical row. The A and the B vari-
eties are arranged in pairs which alternate in the same vertical
row and hence these two should be most fully reciprocating and
also most fully similar in season of bloom, habits of growth, and
cultural requirements. This arrangement would also promote
cross-pollination of any one short-cycle B variety during the
second period of opening with pollen of two other B varieties
standing next in the diagonal rows.
Special means of promoting cross-pollinations in avocados have
been suggested and to some extent employed. Trees of two or
more varieties may be planted close together so that their branches
somewhat interlock and combine to make one compact group.
Another method is to so graft that the branches of two or more
varieties are grown on the same trunk and root system to form
one tree. This has been done to some extent both in California
and in Florida but the results have not been satisfactory. In re-

Bulletin 257, The Pollination of Avocados 39

porting his experiences with this method the late Wm. J. Krome
stated (see Proceedings Florida State Hort. Soc. for 1925, p. 90-91)
that "I have never found that it was a wise thing to do. The
stronger variety will almost every time dwarf the weaker variety,
and you will wind up with only one variety on the stock." The
writer knows of no reports on the results of tests in compact
group-planting but the success of such a method would probably
involve the grouping in one unit of trees that grow with almost
the same vigor.
At the present time definite answers can scarcely be made to
the numerous practical questions which growers of avocados are
certain to ask regarding interplanting. Yet enough is known to
attempt statements which seem most safe and reasonable both
in regard to the known facts and to the means which are most
likely to lead to the best, the quickest, and the least expensive
solution of the pollination problem in the culture of avocados.
Will all varieties be benefited by proper interplanting? Every
clonal variety and every seedling thus far studied in California
and Florida has shown a daily synchronous alternation of dichog-
amy which unmistakably limits and restricts self- and close-polli-
nation. In the sequence of these daily alternations there are the
two main groups from which various clonal varieties may be
selected which afford the most complete and the most remarkable
reciprocations for cross-pollination. Cross-pollination should
greatly increase the chances that fruit will be produced in abund-
ance and with greater regularity. In normal flower behavior and
also during various grades of abnormal flower behavior the rule
is that flowers can not be pollinated during the first period of
opening unless by reciprocating cross-pollination.
To what extent are any varieties self-fruitful? In regard to
this point the writer stated in 1923(3, pp. 42-93) that "Experience
seems to indicate that possibly some varieties may set fruit when
there is no opportunity for cross-pollination" and also the irregu-
lar and off-stride flower action which gives opportunity for dichog-
amous close-pollination was fully described.
A later report(15, p. 84) published after the studies of 1925 in
Florida may be quoted as follows: "Thus the setting of fruit by
isolated trees or by trees of a solid planting of one variety and
particularly by trees abundantly worked by bees in tenting experi-
ments may be expected. It is possible that a peculiar set of local
weather conditions may sometimes favor setting of fruit year

40 Florida Agricultural Experiment Station

after year without cross-pollination. Some varieties such as the
Trapp may have a flower behavior that enables them to be more
self-fruitful than are other varieties. But even for such varieties
there is no doubt that a proper interplanting will increase the
chances of many more proper pollinations and to this extent insure
the production of more uniform crops."
There seems to be considerable evidence that at least certain
varieties may be self-fruitful, especially along the coastal region
south of Los Angeles in California. Observations and tenting
experiments at Point Loma have been interpreted by Clark(1, 2
and 3) to signify that certain varieties are self-fruitful in that
location. In a recent statement Dr. J. Eliot Coit remarks as fol-
lows: "However, so many other factors enter into the question
that throw the regular rhythm out of stride that the practical
growers I know who have been watching this question of cross-
pollination over a period of years have come to the conclusion
that these other factors are more important than the normal
opening or closing time of the blossoms. I can take you to any
number of groves where solid Fuerte or Taft plantings standing
alone produce heavily. One particular instance is three Taft trees
where there are no other avocado trees near for a long distance
that are outstandingly heavy and regular producers. Therefore,
the tendency of the growers today is to go ahead and plant the
varieties they want to plant that will produce most money for
them and trust the trees for their own pollination" (Year Book,
California Avocado Association for 1931, p. 111).
In answer to inquiries regarding the pollination of avocados
J. G. France, Farm Advisor of San Diego County, California,
makes the following published statement: "I hope we are going
to find out after a while. There are some funny things about it.
Some people have Fuertes and nothing else and have good crops,
and some have everything else and have poor crops, and then
vice versa. Without a little more careful observation, I can't
even hazard a guess. I have an opinion, though. I'm inclined to
think that temperature conditions are more important at the time
of setting than having a number of avocado trees around" (Year
Book, California Avocado Association for 1931, p. 112).
The statement quoted above that temperature is an important
factor in fruit setting is pertinent both as to the effect on flower
behavior in relation to pollination and to the effect on the processes
of fertilization after pollination. Under what may be called the
more favorable temperatures flower behavior is more regular and

Bulletin 257, The Pollination of Avocados 41

complete in the alternation and there is less opportunity for close-
pollination. Without doubt there are temperatures which prevent
fertilization to any kind of pollination. Those temperatures which
tend to.favor the processes of fertilization after self- and close-
pollination can be assumed also to be more favorable to fertiliza-
tion after proper cross-pollination.
Robinson(7) has recently surveyed the evidence for self-fruit-
fulness and considered the flower behavior that may be involved,
and he concludes that "Despite the observed instances where
certain avocado varieties are evidently not dependent on cross-
pollination for fertilization of the flowers, it is believed that under
most circumstances interplanting of reciprocating varieties, to-
gether with the use of bees as pollinating agents, provides a worth-
while measure of security against faulty pollination."
According to present knowledge the self-fruiting of any avocado
which has seeded fruit depends (1) on dichogamous close-pollina-
tion during an overlap of sets or when there is the forcing of
flowers by insects, or (2) on self-pollination of flowers whose
pistils remain receptive until pollen is being shed from their
The writer is inclined to believe that flowers in abnormal open-
ing are, as a rule, unable to function in any relation. Yet rather
slightly abnormal cycles produced by moderately inclement
weather, especially if it be long continued, may promote self-
fruiting especially when insects are active. It also seems to the
writer that the forcing of flowers and the self-pollination of cer-
tain flowers by insects during normal opening will be more liable
to lead to self-fruiting. On account of the shorter cycle of dian-
thesis it would appear that self-pollinations are more likely to
yield fruit on B than on A varieties except when their flowers are
open in the long cycle of 54 hours.
How to make avocados bear adequately and regularly has long
been and still is a matter of concern to avocado growers both in
California and in Florida. The erratic and poor bearing of certain
varieties has been a chief factor in their being discarded from
general culture. The flower behavior of all varieties clearly indi-
cates a remarkable adaptation for cross-pollination which can
best be provided for by proper interplanting and by supplying
honey bees in abundance.
To ignore this condition by planting in solid blocks invites low
yields of fruit. This means financial loss to growers especially
during the period when the relatively few most self-fruitful vari-

42 Florida Agricultural Experiment Station

eties are being discovered and the areas are being delimited in
which the environmental conditions favor or induce a type of
flower behavior that is most favorable to self-fruiting.
It is without doubt advisable (1) to interplant avocados on the
basis of their flower behavior, (2) to supply bees in abundance
to effect pollinations, (3) to make special experimental tests of
various types of interplanting with different combinations of the
most promising varieties, (4) to make definite studies of existing
commercial orchards regarding the yields of each tree and the
disposition of the different varieties, and (5) to continue careful
studies of normal and abnormal flower behavior, of the means of
pollination, and of the processes of fertilization.
The combined results of such efforts should rather quickly and
definitely reveal what, if any, varieties are self-fruitful and under
what conditions and to what type of flower behavior and pollina-
tion they thus operate. The results thus obtained should also
reveal to what extent varieties require cross-pollination and what
particular combinations of these are most favorable to fruit set-
ting. Perhaps it is not too much to hope that in time a group
of ideal avocados may be discovered or developed by selective
breeding which among other desirable qualities are fully self-
fruitful. That such varieties are not now in existence among the
varieties commonly cultivated is evident. Until such varieties are
definitely known it is a safe and wise plan to interplant on the
basis of flower-behavior and with due consideration of other qual-
ities and cultural requirements.
There appear to be few legitimate objections to the interplant-
ing of avocados. In making interplantings along the lines that
now seem most promising the grower has nothing to lose and
there is the chance that he will profit by increased production.

This bulletin is a report chiefly of the investigations which the
writer made in Florida in 1925 and 1932, but to some degree the
conclusions and the data involve the earlier studies in California
during 1923.
The investigations in 1925 were made under the auspices and
with the financial support of The New York Botanical Garden
and the Farm Bureau of Dade County, Florida. During most of
that period of work E. M. Savage of the Bureau of Plant Industry,
U. S. D. A., participated in the investigations as did also T. Ralph
Robinson for a part of the time. Valued assistance and coopera-

Bulletin 257, The Pollination of Avocados 43

tion were rendered by various members and officers of the Florida
Avocado Association and especially by the late Wm. J. Krome and
by J. S. Rainey, county agent of northern Dade County.
The investigations of 1932 involved cooperation between the
Florida Agricultural Experiment Station and its Sub-station at
Homestead, The University of Miami, The Board of Commission-
ers and the Farm Bureau of Dade County, The New York Botan-
ical Garden, and various growers of avocados. The services of
the writer were provided by The New York Botanical Garden and
the necessary expenses involved were provided by the Board of
Commissioners. The work was organized chiefly through the
interest of Professor Alfred H. Gilbert of the University of Miami
and his helpful cooperation continued throughout the period of
work. Under a fellowship grant from the University of Miami
T. W. Young collaborated especially in the technical studies of
fertilization, and E. Mortin Miller of the Department of Zoology
conducted studies of insect activities and made important findings.
The Sub-station at Homestead was the headquarters for the work
and afforded ample facilities for it. Especially to Dr. H. S. Wolfe
in charge of the Sub-station and to Director Wilmon Newell, As-
sistant Director H. Harold Hume and Dr. A. F. Camp of the Ex-
periment Station the writer is indebted for interest and coopera-
tion in various features of the work. C. H. Steffani, county agent
of southern Dade, cooperated especially in arranging for meetings
and contacts with avocado growers. The field work was done
chiefly in the avocado groves of Mrs. Isabella Krome, W. K. Wal-
ton and those under the management of H. E. Schaff-all of whom
gave hearty support to the endeavors. To all these and to numer-
ous others not here mentioned by name the writer extends his
thanks and appreciation.
This bulletin draws freely from earlier publications of the writer
and especially from the report on "The Flower Behavior of Avo-
cados" published by The New York Botanical Garden in 1927.
The illustrations of figures 1, 2 and 3 are from photographs used
in that report and the plates of Figures 4, 6, 7, 8, 9, 10 and 11 are
of those made for that report and here used by permission. This
repetition is, it is felt, justified, first because a large number of
the separates of that report which were printed for distribution
to avocado growers in Florida were destroyed in Miami during
the hurricane in 1926, and, second, because these photographs and
charts very fully illustrate the matters involved.

44 Florida Agricultural Experiment Station


1. CLARK, ORANGE I. Avocado pollination and bees. Annual Report, Cal-
ifornia Avocado Association, 1922-1923.
2. Avocado pollination tests. Annual Report, Califor-
nia Avocado Association, 1923-1924.
3. CLARK, ORANGE I., and CLARK, ABBOTT B. Results of pollination and
other experiments on avocados at the orchards of the Point Loma Home-
stead. Annual Report, California Avocado Association, 1925-1926.
4. HODGSON, ROBERT W. The California Avocado Industry. Circular 43,
California Agricultural Extension Service, 1930.
5. NIRODY, B. S. Investigations in avocado breeding. Annual Report,
California Avocado Association, 1921-1922.
6. ROBINSON, T. RALPH. Pollen sterility in the Collinson avocado. Jour.
of Heredity 21: 35-38. 1930.
7. Some aberrant forms of flower mechanism in the
avocado. Year Book California Avocado Association for 1931.
8. ROBINSON, T. RALPH, and SAVAGE, E. M. Pollination of the avocado.
Department Circular 387, U. S. Dept. of Agriculture. 1926.
9. SKUTCH, ALEXANDER F. Observations on the flower behavior of the
avocado in Panama. Torreya 32: 85-94. 1932.
10. STOUT, A. B. Avocado studies: pollination and setting of fruit. Farm
and Tractor Section, Los Angeles Sunday Times, April 29, 1923.
11. Clocking the avocado-A study in cross-pollination.
Los Angeles Sunday Times, July 8, 1923.
12. A study in cross-pollination of avocados in southern
California. Year Book, California Avocado Association for 1922-1923.
13. The Flower behavior of avocados. Mem. N. Y. Bot.
Garden 7: 145-203. 1927.
14. Dichogamy in flowering plants. Bulletin Torrey
Bot. Club 55: 141-153. 1928.
15. STOUT, A. B., and SAVAGE, E. M. The flower behavior of avocados with
special reference to interplanting. Proc. Florida Hort. Soc. 1925: 80-89.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs