A study of the gases of Emmental cheese


Material Information

A study of the gases of Emmental cheese
Series Title:
Bulletin / U.S. Department of Agriculture, Bureau of Animal Industry ;
Physical Description:
32 p. : illus. ; 24 cm.
Clark, W. Mansfield ( William Mansfield ), 1884-1964
Govt. print. off.
Place of Publication:
Publication Date:


Subjects / Keywords:
Cheese -- Analysis   ( lcsh )
Swiss cheese   ( lcsh )
federal government publication   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references (p. 32).
Statement of Responsibility:
by William Mansfield Clark ...

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 029612145
oclc - 22233338
lccn - agr12001325
lcc - SF623 .B14 no. 151
System ID:

Full Text

!~-4AX 0o

t q_. . .


1 41

..,[]~~ ~~~~~ %J ............................


ir PWJ
'4 -0 OW14

, tril1,

.1 -r
I I ~1

'II J~
I ~. I'

I -
* 1 -

*1 ~
I4~ 'i$'
I *1
A .*

4 K
It. 14
I. j--.
'A -.
L **~ *~-* 4





a~ r
J~lI -


~ 4 -

.. --*-- -jr -^ ;' ^1..=*:.
~ - ,, .. ^ .,.. -. *:. ,. ,.. ^ *- .. .. ,:.. .-.. . ., :." .
-,.. -- P -; '. .y : .'i
,..- .. , .'. '.-
-- . .. .. *....

r- *, *. 4 ,' T *. /( ,'' / : 1 ., ".. *-..

S. ... . ... . q. .
$ ,.,*. . ':: , ,-. -*. ', ..
.. -, _o .. r " A """" ." I "

.- ; i., *.f - . , .. * .* .., ' *... . ,' : '-:. '. ... ... ,: .* ";.'*.

A. 'A -
r*'- 'S *t 4VTY --- L4I'^Jt
ll' \ v

--,. .~ *~ Add~f.. rtdi
.. ..r. ;....

-I - i .. . . . a. ,' .. .

"r." -
.0s. A; .. ; . /.... ..
-, , ,I / "^ i * " r, ": *. "* "* [ '' a .- : .< * *S a
', . .. .. ,; .. r", " " "< 'r ". ^ ,. .- .. ",y ,"- A "", l

j - -. ," '* ", ". "... ", *, .. ... ...,- .- ; "
f .. . .- - .. ....... . u .. *> . ... ," :..: ..
. -, ,* -,./ ,*'t " ." ,., ;" .- T '.'.: 1,
.,* * -. -,, ,;\.- ; . .. ...:' -: ".* ; * * *

.. . ,. ; ,. ,. .,. : 34.
o. l .* ; ,: ,., :. *., ..... ,* .*- l :.-^ *?-. i -,
.* ; .;. , .. < ..: . '. . ..* .; : ; *. * .:.*:. -.., -
,* ,* -. ... .. .. . " .. .;. .'-. ; . .. ^ ...^ f ^ .^
'- -"- . I- .-- .,"* ".- "
-. .... .. .- ) .. .. ..-'*'l ., t 7 / ..-- ...^...^ .... q'.
-' ... :' **......" } *: ... . -*:? .,,. ... ', .. 4 .,. ^ :v. -..:' ^

N *: d ''
,"" : *.dam
1 f< .* ". :'" '^ .- 1 ,: ". ": ^ '.:'. -"^ : .1:-:,

*- .' : '" "".- '; '>,'- "'" . '*'"* ; "; -^ ^ ."1
*. .' ...* .. ... ... . ... v .. ?, ,.. -..^ ,-
": .* '* *, "" *" ./ "* ./-' '* "- .. ..... '- ". ":'-,-^ ^
I - . "* .. : ,- t .. .*"

; ,- ": ,, :. :." : ,.* . .... ,,'. ,\ >" ., ^,..., ..{ ,- .? .^
"- ', .. ,* -*,. : ) , -..- : .. 4' ,f^ *. 4 .
." ,* ,' .' / ":- . t .f l . I r ... ; a
', ., ... r -. ,- *, .'" ... ; .. ..
/ .. . "- -, '., ." ", ... :,^ .. . *... **.'" ." ,, .; "% .":
.. :.. * " , . .....-* : .. ." ^ ... ., ... .. .- .^ .;. :i : ,
" .* . v . ', "., ". ";. *.; ** .* ^.'. v'^:"

.. . .. ;- .- .- *: : ' 1, "] .: ...'.";-" .^. .3 S. .
y '.. i. ... . .. .* . . .* .** :.. 1 ^ .. ..1
,- .* ,: " .. . .".' ::* : ". *' *,; t "",
.. -i ~ .: .* *,. ..* "-,. .,,... .** ** **
* "-. ;' ' ." ", "^ 1 .^* .' ; : ; ; > : E : ., ,
.. ,. .-.^ : ,-.. ,.. r. r'. .':i, .
.* '*..... ,. ,l,

-" ," * ' : -" .' i f .." r i" './ ,;.- , * :*r 'i ". ",: 7*:t ~
* r ** . .. ... .- d ^ . ... .. ... ,
'", :',,-, .- It-,' i

.. 9 .. t .4-. ." ...
.*1 c, & *,,,1, ,- .v ?, g.".

.1 2 -.. ... .tt. "* "t ,' ." '
.-, ,4 .. .'. .. ." &

, .. t . A r ,

,.It..,, I ; i % ... .

...1.". ,, 1. . ", ,,

.7..1 # 1 ". . -. . . ." :,
I.'. I2-. :'


"* j *

Imued eptrmber 7. 1912.



A. I). MEl-VIN, MCiisr oru IuRaJ.









Chemist, Dairy Dizision.

IB "1

& 0




PH. D.,

i A


Chief. A. D. MELVIN.
Assistant Chief: A. M. FARRINGTON.::
Animal Husbandry Division: GEORGE M. R6MMEL, chief.
Biochemic Division: M. DORSET, chief.
Dairy Division: B. H. RAWL, chief.
Field Inspection Division: R. A. RAMSAY, chief.
Meat Inspection Division: RICE P. STEDDOM, chief.
Pathological Division: JOHN R. MOHLER, chief.
Quarantine Division: RicHARD W. HICKMAN, chief.
Zoological Division: B. H. RANSOM, chief.
Experiment Station: E. C. SCHROEDER, superintendent.
B. H. RAWL, Chief.
HELMER RABILD, in charge of Dairy Farming Investigations.
S. C. THoMPSON, in charge of Dairy Manufacturing Investigations.
L. A. ROGERS, in charge of Research Laboratories.
ERNEST KELLY, in charge of Market Milk Investigations.
ROBERT McADAM, in charge of Renovated Butter Inspection.
2 "i

ADDITIONAL COPIES of this publication
.ffmay be procured from the SUPERINTKND-
mT or DocuxMENTs, Government Printing
Office, Washington, D. C., at 5 cents per copy

*l I
MeatInsectin Dvisin: ICE SEDDO, cief
Path~ogcal iviion:JOH R. OILR, cief

Quaanin Dviio: ZC R W HKMNchef


W.ashington, D. C., April 23, 1912.
'f SIR: I have the honor to transmit, and to recommend for publi-
i cation in the bulletin series of the bureau, the accompanying manu-
S script entitled "A Study of the Gases of Emmental Cheese," by Dr.
S William Mansfield Clark, chemist in the Dairy Division.
The so-called "eyes" in Swiss cheese are, as is well known, its most
prominent characteristic, and its commercial value is largely depend-
ent upon the proper size and spacing of these eyes. Furthermore,
much depreciation in the value of this popular variety of cheese, in
both the domestic and foreign kinds, is known to exist because of
defects in eye formation. The experimental work herein described
concerns the chemical contents of these eyes, and although consider-
able work has been done in Europe with the object of discovering the
cause of eye formation, there has hitherto been no investigation made
of the gases which are immediately concerned in the process. Dr.
Clark's studies are therefore calculated to be of value to the scientific
as well as the practical side of the industry.
Chief of Bureau.
Secretary of Agriculture.







Introduction-.............................................................. 7
Description of apparatus and methods of collecting the gases.................. 9
Method I.......................................................---------------------------------------------------...... 9
M ethod I I ............................................................ 10
Method of analysis....................................................-----------------------------------------------..... 11
Discussion of the analyses...............................................------------------------------------------... 12
Absorption of oxygen....................................................--------------------------------------------- .1...
The permeability of cheese to gases------------------------------------......................................... 20
Nitrogen dissolved in curd---------...............--.---..--------------------------............ ....... 24
Does nitrogen originate in situ?.......................................... 25
Relation between carbon dioxid and volatile acids................... ...... 26
Sum m ary-.................................................................. 31
References to literature........................................----............--- 32

:i::.i iu



FIG. 1. Apparatus for collecting gas from the eyes of Swiss or Emmental cheese. 8
2. Apparatus for pumping gas from cheese........ ......---.----.--.--.-- 10
3. Apparatus for studying the absorption of oxygen by cheese ........... 14
4. Device for ascertaining permeability of cheese to gases- ................ 20
5. Apparatus for determining amount of nitrogen in curd................ 24



4ft The "eyes" of Swiss or Emmental cheese are its most striking
^H^^ ~characteristic. Their formation is a fascinating subject to the bio-
,,,:S logical chemist, because of a supposed localization of reactions gen-
erating considerable quantities of gas, and because of the produc-
ft!11 tion of a plasticity among the colloids of the cheese, which makes
M possible the peculiar mold of the cavities.
:l : To the cheese maker the formation of the "eyes" is a matter of
great importance, since their size and proper spacing determine in
.f large measure the commercial value of the cheese. In certain dis-
tricts of Wisconsin visited by the writer the dealers rely almost
entirely upon these features, and, shortly after the eyes have reached
their proper development, relieve the maker of further care. The
American makers of Swiss cheese are, therefore, unable to attend
to their cheeses in that mellow old age upon which so much of the
fine flavor of a true Emmental cheese depends. However much this
quick marketing is to be deprecated, the fact remains that it raises
the relative importance of the eye formation and adds significance
to whatever knowledge can be gained concerning the process.
Some years ago BAchler,1 a cited by Jensen," estimated that 25
per cent of the cheeses made in Switzerland were considerably reduced
in value because of imperfect eye formation. How far this enormous
loss has been lessened in recent years as a result of scientifically con-
trolled manufacture can not be said, but in this country, where large
numbers of Swiss are still using the antiquated methods of their fore-
fathers, BAchler's estimate is probably not too high. The wide dif-
ference in market price between domestic and imported Swiss cheese
bears out this statement.
Considerable work has been done in Europe in the effort to uncover
the cause of eye formation, and, through the labors particularly of
a The reference figures relate to the list of references to literature at end of bulletin.

.:Em'.m":'; ii:


Von Freudenreich and Jensen, a well-founded theory has been proposed :
which will be discussed later. No one, however, has made a study of
the gases which are themselves the immediate cause of the eye forma,-
tion, and it was with the hope that such a study might furnish valu-
able data that the research herein described was undertaken. 11
nothing more is demonstrated than the composition of the gas in:



..: -........ -........ **** --
FIG. 1.-Apparatus for collecting gas from the eyes of Swiss or Emmental cheese.

the eyes, this alone justifies the work, for the extensive researches on
the eye formation in Emmental cheese have led to but one conclu-
sion that can be called positive, and that is that a final explanation
will be reached only when every phase of the subject has been sub-
mitted to exact quantitative study.

.. illal" /GASES.

S The collection of the gas in the eyes by cutting the cheese under a
*N' bell jar filled with water, as was done with Edamn cheese by Boekhout
and Ott de Vries,2 is a simple and valuable method, but one which
is hardly to be called accurate, owing to the high solubility of certain
gases in water. In place of such a method an apparatus was devised
for collecting the gas over mercury. This is shown in figure 1, the
S" procedure being as follows:
I The glass cylinder A is forced a short distance into the body (if
I the cheese until it is firmly held. It is then clamped in position.
.. Around the outside the cheese is cut away sufficiently to leave a
iii channel into which mercury moistened with mercuric chlorid solution
W. is poured. This forms a seal preventing entrance of air. The head
j*r of the shaft B is now resting on the surface of the cheese. Through
liii its capillary mercury is run into the cylinder, displacing the air until
NJ1 it finally runs out of the side arm D and up through the annular
space between the shaft and the shoulder of the cylinder. The short
length of thick rubber tubing at E is then very tightly bound with a
: rubber band, leaving mercury in the small cup above, and thus
S effectually closing this opening against the entrance of air. When
the cylinder and side arm are thus completely filled with mercury, a
i' receptacle filled with mercury is brought over the end of the side
S arm (in a mercury trough, of course) and serves to retain the col-
S elected gas until the time of the analysis. After these preparations
the shaft is pushed down into the cheese. When it punctures an
eye this can readily be felt. Since the head of the shaft is larger
than the shank, there is left an annular space for the escape of the
gas. This gas is displaced from the eye partially by the mercury of
S the cylinder, which finds its way to the lower level, but more largely
by the mercury which runs in through the capillary in the shaft. The
exit of this is prevented from becoming clogged with cheese by care-
fully blowing it out just behind the head, as shown in the diagram.
When the gas is displaced from the eye it is displaced from the
cylinder into the receiver by continuing to run in mercury through
the shaft from the reservoir C. Between this reservoir and the shaft
is placed a bulb which prevents the mercury from sweeping in bub-
bles of air.
In the samples of gas collected with this apparatus seldom was more
than a trace of oxygen found. This in itself shows that the gas was
obtained without contamination by air.
422080-Bull. 151-12----2



For the collection of gas from pinholess" the foregoing apparatus
was of little use except in one instance to be mentioned later. To :
collect the gas from this form of hole, as well as the gas in the body .]
of the cheese, the apparatus shown in figure 2 was used, as follows:
Samples of cheese taken with a trier were introduced into the ;.
glass cylinder A. The rubber
Jj Stopper at B, attached to the '
c mercury vacuum pump with or J
without the intermediate connec- 1
c tion C, was forced in securely and I
protected from leakage by the
mercury seal. Upon raising the J
/? leveling bulb D the cheese was |
flooded with mercury and the sur- ;
rounding air was forced over into i
the pump until the mercury stood
at the stopcock E. To prevent |
bubbles of air being trapped under
the cheese the lower ends of the
A plugs were sharply beveled. Bub-
bles of air of course adhered to the
rough surface of the cheese and its
smaller exposed cavities. This
aff error is inherent in the method,
but was reduced by suddenly drop-
ping the leveling bulb with the stop-
cockE closed, and then driving the
| air, which had expanded into the
Vacuum, past the open stopcock.
i|The glass tube with its trap
S |which connects A with the level-
| ing bulb was made sufficiently
Long so that D might be lowered
the barometric distance below A,
^y ~and thus leave the cheese exposed
FIG. 2.-Apparatus for pumping gas from cheese, and thus leave the cheese exposed
to a fairly high vacuum even be-
fore the pumping commenced. After exhausting the pump up to 0
E this cock was opened, and the gas pumped from the cheese and AI
delivered into a receiver. !
The mercury pump used in this as in other operations to be de-
scribed later was Antropoff's modification of the Topler. A full
description of the pump and its appurtenances will appear in the
account of another investigation.
4 ::




~ The gas was analyzed with a special sot of biuretts. and pipettes
r designed for the analysis of small quantities of gas produced by
bacteria. A few of the first analyses were miiadn with a. burette spe-
d&ially designed for volumes as low as 0.5 c. r. In atll the analyses
the confining liquid was mercury, and use was made (if a device for
I!i extremely accurate separation of gas from absorbent.
~: Thirty-three per cent potassium hydroxid solution, in quantities
I' appropriate for the volume of gas analyzed, served as absorbent for
S carbon dioxid. Hydrogen sulphid, after preliminary qualitative
tists, was assumed to be absent, although it is of course l)possible
that, if present originally in the gas, it may have been taken up by
the mercury. That any of this gas occurs in the eyes is, however,
S very improbable, for its odor was never detected. For hydrogen
; sulphid and mercaptans the nose is many times more sensitive than
i is the spectroscope for sodium,6 and unless the other and milder
odors of Swiss cheese exercise a surprisingly intense hindrance to the
detection of hydrogen sulphid and mercaptans we may justly say
That these vapors were absent. With Nessler's reagent very slight
traces of ammonia were detected. For oxygen alkaline pyrogallol
or long-continued contact with phosphorus was used. Combustible
gases were estimated in several ways. Explosion with oxygen, in
the presence of electrolytic gas when necessary, was used in several
instances. For one case combustion with a platinum sponge was
tried. For the small percentages of combustible gases found the
method of Dennis and Hopkins 5 was found to be the most satis-
factory. This consists, essentially, in leading the gas slowly into a
measured volume of oxygen and there burning it slowly and quietly
with a platinum wire heated by an electric current.
TABLE 1.-Analyses of gas collected by puncturing apparatus from eyes of Swiss (Em-.
mental) cheese-3Method I.

S Contraction- Composition.
Desig- Total
tol- Due to Due
ume of Due aor D to u Description of cheese.
fe- tidn so corn- Cot. O. car- H2. N2.
"eite& wt tion bus- bonr.-
with for 0. bos bons.
KOH. tion.

Per Per Per Per Per
C. c,. C. C. c. C. c. cent. cent. cent. cent. cent.
a 0.96 0.55 .............. 57.3 .............................. Imported, eyes normal.
b 2.73 2.29 0.00 ...... 83.9 1 0.0 ....... ........ .. ... Do.
0 1.66 1.11 .01 0.10 66.9 Trace. (?) 4.00 29 Imported, eyes (?).
d 4.77 2.44 .02 .02 51.2 Trace. 0.0 Trace? 48.8 imported, eyes normal.
I 1.25 1.00 .00 .00 80.0 0.0 0.0 0.0 20.0 Do.
f 3.44 2.23 .00 .60 64.8 0.0 0.0 11.6 23.6 Domestic eyes thickly
S4.02 2.52 .........45 62 .......... 0.0 7.5 29.8 f crowded.
g 15.24 13.77 .00 .76 90.4 0.0 Trace? 3.3 6.3 Do.
h /7.56 6.14 .02 .43 81.2 Trace. 0.0 3.7 15.1 Jtmported eyes thickly
S 4.99 4.04 03 .30 80.9 0.6 0.0 4.0 14.5 crowded.
114.47 12.91 .03 .00 89.2 0.2 0.0 0.0 10.6 Excellent imported, eyes
9.99 8.95 .01 .09 89.5 Trace. 0.0 0.6 9.8 very regular.
J 5.42 3.04 .02 .06 56.0 Trace. 0. ? Trace? 44.0 Imported, large hole.
39.61 4.96 2.35 .01 3.63 47.4 Trace. 0.01 48.8 3.8 Very gassy in press.


The analyses of the gas collected by Method I are given in Table 1, !
and of that collected by Method II in Table 2. All volumes are for
0 C. and 760 mm. When the gases were collected from a cheese pro-
cured at the market, a sufficiently large slice was purchased to pro .
vent undue exposure of the eyes, and this was carried immediately
the short distance to the laboratory, and the gas at once collected.
In most cases the shaft punctured or grazed more than one eye, so
that the analysis gives the true average for several eyes.

TABLE 2.-Analyses of gas collected by pumping from Swiss (Emmental) ehees-
Method II.

Time Total Weight Amount Composition.
Time Total c of gas ______________
os pf gas col- cs per 100 Description. of cheese.
ing. elected. ed grams of
cheese. C02. H0. 22.

Per Per Per Per
Hours. C. c. Grams. C. c. cent. cent. cent. cent.
3 20 2.36 .......... .......... 76.3 1.7 0.0 22.0 Almost blind. Several small
holes, either pinholes or in-
hibited eyes.
39-45 ........ 2.31 .......... .......... 77.5 2.6 0.0 19.9 Do.
39-11-2 20 6.41 .......... .......... 80.8 2.0 0.0 17.2 Do.
46-4-1 ........ 3.20 50 6.40 50.6 1.0 0.0 48.4 Do.
W2.. 20 13.60 53 25.7 84.5 2.2 0.0 13.3 Fine domestic cheese just be-
ginning eye development.


If the values obtained in this study of the gases found in the eyes
of Swiss cheese are compared with the values obtained by Boekhout
and Ott de Vries 2 for the gases in Edam cheese, it is seen that the
latter obtained much lower percentages of carbon dioxid and corre-
spondingly higher percentages of nitrogen. The explanation becomes
apparent when it is remembered that Boekhout and Ott de Vries
collected the gas over water, while in this investigation it was collected
over mercury. The two methods were compared in the case of cheese
h, as follows:

Method. COg. 0,. Hu. N,.

Percent. Percent. -Percent. Percent.
Collection over mercury.... ............................. 81.2 Trace. 3.7 15.1
Collection over mercury... ............................... 80.9 0.6 4.0 14.5
Collection over water........................................... 34.8 1.9 1.9 61.4

This result is what might have been expected, namely, an absorp-
tion of much carbon dioxid and a little hydrogen by the water, and,
in return, an increase in the amount of oxygen as well as an increase
in percentage of nitrogen. Boekhout and Ott de Vries have them-
selves called attention to this, and claim only qualitative value for
their results. The types of holes from which they isolated gas were
small cracks corresponding to the Emmnental "riszler," small round
holes, and large cracks termed "knijpers."



il! Qualitatively the composition of the gases was the same, namely,
ji carbon dioxid, hydrogen, nitrogenpand oxygen. (f these they elimi-
n ated oxygen as due to contamination. In the case of the "knij-
SpesM or large cracks, 52 to 249 c. c. of gas, wn,('r collected instead of
5 to 22 c. c. as in the case of the smaller holes. Assuming that the
same volume of water was used, we would expect a truer value to be
obtained for the analysis of the larger volumes, in which case the
attention is struck by the large percentage of hydrogen. The signifi-
"cance of this will become apparent when tfle results (Ili Emmental
cheese have been assembled.
: It is clear from the analyses of gas found in Emmental cheese
'I that carbon dioxid and nitrogen arc the chief constituents of tlhe gas
Found in normal eyes. The oxygen in most cases is hardly more than
II would be expected to come from the minute bubbles or surface
S layers which adhere to the glass walls of the apparatus. To what
i gas the contraction after explosion with oxygen is to be ascribed is a
1 difficult question to settle. In some cases, where the contraction
S was sufficiently large to justify further absorption with potassium
hydroxid, the absence of any further contraction in volume justifies
the conclusion that the combustible gas was chiefly hydrogen. In
other cases the small contraction might have been due to any one of
S a number of gaseous combustions.
For further information it was decided to examine specimens of gas
spectroscopically. The gas freed from carbon dioxid and possible
oxygen was passed over phosphorus pentoxid into a dry, exhausted
Plficker tube. The discharge of an induction coil was then passed
between aluminum terminals, and the spectrum observed with a
prism spectroscope. At the same time comparison was made with the
spectrum of a similar tube containing pure hydrogen. Minute traces
of hydrogen are to be expected when metal terminals are used, but,
with the low resolving power of the spectroscope employed, the nitro-
gen spectrum so obscured the possibly present red line of hydrogen
that it was not observed with specimens of pure nitrogen. A known
Sample of nitrogen containing about 0.05 per cent of hydrogen gave
a brilliant hydrogen spectrum, whose intensity could be made more
sharp at the expense of the nitrogen spectrum by suitable varying of
the pressure.15 The recognition of 0.05 per cent of hydrogen was
therefore assured.
A small experimental cheese, which had begun an apparently nor-
mal eye formation and then ceased entirely, was pumped out by
Method II and its gas submitted to spectroscopic examination.
Slight evidences of hydrogen were observed. Samples of gas taken
from cheeses which yielded 3 per cent of combustible gas gave very
brilliant evidences of hydrogen.




In samples of gas taken from the normal eyes of two cheeses pur-
chased on the market no hydrog&h line was observed, nor -was the
hydrogen spectrum observed in the gases of a normal cheese evolved
during the period of its maximum eye formation.
These results, though not extensive,
[are sufficient to show that hydrogen
D plays no r61e in the formation of normal
eyes, provided we assume that any hy-
c drogen formed has not escaped collection
by rapidly diffusing through the cheese.
H To make sure of this point the following
experiments were conducted:
Two cheeses purchased in Wisconsin
were found to be developing normal eyes.
These eyes, though too thickly scattered
Sfor the modern market standard, would
6 have been declared typical some years
ago. When each cheese was apparently
at the height of its eye formation, plugs
were taken, and introduced into the tube
A, figure 3, without that part illustrated
at the side and lettered G, F, and E. To
A4 guard as far as possible against infection
E_ in transference the trier was flamed, and
the tube was sterilized at 170 C., with
cotton plugs at B and C. After intro-
ducing the plugs of cheese they were fol-
lowed by the flamed cotton plug and then
-- a rubber stopper dipped in hot rubber
: '< cement. The stopper was forced in and
"^'"J ~held in place till the cement a had cooled,
When several layers of the same cement
were added to the exterior. This made
S a thoroughly gas-tight seal. The capil-
FiG. 3.-Apparatus for studying the ab- lary end was now attached to the mercury
sorption of oxygen by cheese. pump by means of securely tied rubber
tubing completely covered with a mercury seal. Then the tube was
Forty-six grams from one of the Wisconsin cheeses were exhausted
for two hours, during which time it continued to give off small
quantities of gas. The pressure was finally reduced to 2 mm. (meas-
ured on a McLeod gauge). The stopcock D was then closed, and the
a The cement was made byheating rosin several dayswith as much fine-grade rubber as itwould dissolve.
Dr. Nutting, of the Bureau of Standards, who kindly furnished the receipt, stated that he had used this
cement in refined vacuum work with entire satisfaction.




tube allowed to remain in connection with the pump overnight.
The next morning the pumping wis resumed, and a pressure of 2 mm.
S again obtained. The gas which had collected overnight amounted
ii to 7.23 c. c., N. T. P. Its analysis follows:
Original volume............................................... 7. 23
Residue after absorption with KOII............................ .17
CO ...................................................... 05
Oxygen added up to....................................... .2.28
Volume after combustion with heated platinum spiral........... 2.26
4Contraction.............................................. 02
The tube was then sealed off in a blowl)ipo at thi constriction II
S and kept for six days at 25 C. To collect the gas from this sealed
S tube the following method was used. The capillary tip of the seal
I was scratched with a diamond, and then pushed up into the tube
I: leading from the pump as at C, figure 3. Connection was made with
i: a rubber tube securely tied and covered with a mercury seal. Having
1i| exhausted the pump up to the tip of the seal, the tube was turned

munication established between A and the pump.
Slightly and sharply. The tip was broken at the scratch, and corn-
The gas thus collected at the end of six days amounted to 10.12
c. c. The tube was allowed to stand connected with the pump over-
Snight, after which an additional 2.75 c. c. of gas were collected.
These two volumes were united and analyzed 99.3 per cent carbon
dioxid. The residue was hardly sufficient to justify further analysis.
It was made, however, and a minute contraction observed, which was
hardly more than the experimental errors of transference.
Forty-five grams of 'the second Wisconsin cheese submitted to the
same procedure as described above gave the following data:
Gas collected on first standing overnight..................... 11.48
Residue after absorption with KOH........................... .22
C02. ......................................................... 11.26
Residue after absorption with phosphorus..................... .22
Oxygen added up to.........................................a 2.41
Volume after combustion with heated platinum spiral........... 2.39
Tube sealed off and incubated six days at 25 C.
Gas collected after 6 days ...... ............................... 9.29
Residue after absorption with KOH............................ 16
CO2 .................................................... 9.13
Oxygen added up to.......................................... a 4. 91
Volume after combustion with hot platinum spiral.............. 4.85
Gas collected after again standing overnight.................... 5.33
Residue after absorption with KOH ........................... Trace.
a This comparatively large volume was made necessary because of the disadvantageous form of the
Dennis-Hopkins pipette used.



In the above analyses the contraction due to combustion was so
small that further analyses to deterinine the products of combustion
Were impracticable. Nor was it necessary, for, even if the contraction
Were due to but one gas, for example hydrogen, the amount was such
that this gas may be said to be without significance in the formation
of eyes. Doubtless the contraction was in reality due to volatile
organic bodies. The above experiments show that when all the gas
from an actively gas-producing region is collected no significant
amount of hydrogen is found, and thereby the contention is refuted
that, in the analysis of gas in the eyes, hydrogen escaped detection
because of its rapid diffusion out through the cheese.
Pains were taken in these studies to make a strenuous hunt for
hydrogen for the following reason: In Emmental cheese there is
what Duclaux has termed the "initial fermentation" during which
the sugar inclosed in the curd undergoes bacterial decomposition.
Several of the earlier workers on this cheese thought it was the gaseous
fermentation of this sugar which caused the development of eyes. If
so, one would expect to find the gas composed of a large percentage of
hydrogen, since hydrogen is a characteristic product in the fermenta-
tion of sugars by bacteria. This deduction is of course not rigid, but,
from our present knowledge of the gaseous fermentation of sugars by
bacteria, it is highly probable.
Jensen 1 in 1898 pointed out clearly that the gaseous fermenta-
tion of sugar must not be looked upon as in any way directly connected
with the production of normal eyes in Emmental cheese. He
found no trace of sugar in a cheese five days old, although the normal
eye formation had not yet begun. This confirms the analyses made
by various authors. Jensen cited Klenze 13 as stating that the sugar
disappears in 48 hours. But, while the sugar disappears rapidly,
normal eyes seldom begin to develop before the eighth day, and reach
the height of their development long after every trace of sugar has
These facts alone demonstrate that the eye formation does not
depend upon the presence of sugar. Additional reason for so believ-
ing is found in the results herein, in so far as the absence of hydrogen
in the gas indicates an absence of gaseous sugar fermentation.
But it also follows from this reasoning that when a gaseous fermen-
tation occurs while sugar is still present in the cheese, hydrogen is to
be expected. Such a fermentation frequently occurs while the cheese
is in press. Fortunately a cheese was obtained (No. 39-61) which was
known to have given marked signs of gas while under press. From
this cheese gas was collected by the previously described Method I,
with the following analysis:
Total volume of gas collected ................ Cubic centimeters. 4.96
Residue after absorption with KOH.......................do-.... 2.61
CO _,................................................do.... 2.35
Residue after absorption with phosphorus.................do .... 2.60



Oxygen added up to ........................ Cubic centimeters.. 6.14
Volume after combustion with platinum Nspiral............ do.... 2.51
Contraction......................................... do .... 3.63
Residue after absorption with KOH(....................... do.... 2.51
Hydrogen............... .......................... p.Tr vent.. 48.80
Upon attempting to make a second punctureo the mercury broke
through into the hole previously made. The choOso was then opened,
and found to be so spongy that the walls separating the individual
cells were very thin-too thin to withstand the weight of mercury.
To obtain a second sample of gas for confirmatory analysis recourse
was had to Method II of collecting gas, previously described. A high
percentage of hydrogen was again found.
In the further study of this case 52 grams of the cheese wore intro-
duced into the vacuum tube described on page 14 and evacuated to
1 mm. pressure. There collected overnight 7.84 c. c. of gas.
Total volume................................................. 7.84
Residue after absorption with KOII............................ .28
Residue after absorption with phosphorus ...................... 27
Oxygen added up to......................................... 1.08
Volume after combustion with platinum spiral................. 99
Contraction .............................................. .09
The tube was then sealed off and kept nine days at 25 C. Upon
opening it and pumping out the gas by the method previously
described 7.49 c. c. of gas were collected. The residue after absorp-
tion with potassium hydroxid was only 0.07 c. c.
It is therefore apparent that the production of hydrogen, which
was very active while the cheese was in press, had soon ceased, pre-
sumably with the disappearance of the sugar.
The occasional occurrence of hydrogen in small percentages, as
-shown in the table, generally accompanied eyes which in the writer's
judgment were not typically normal. They were either crowded and
distorted or associated with numerous pinholes. It is not, perhaps,
incorrect to say that in all probability there had occurred in these
cases a slight initial gaseous fermentation of the sugar, with the pro-
duction of hydrogen which lingered to contaminate the gas of the
normal fermentation.
An extremely interesting observation was made in the case of cheese
i. (See Table 1, p. 11.) This was an excellent imported cheese with
large and perfectly rounded eyes, well spaced in a body of fine texture
and flavor. In the first analysis of the gas from these eyes no trace
of a combustible gas was found. The second analysis gave 0.6 per
cent of hydrogen. Upon exposing the eyes punctured it was
observed that a slight crack extended to within a centimeter of one of
the eyes punctured on the second collection. This crack was found
to lead directly to a hole some 2 cm. in diameter, the irregular and
apparently corroded walls of which proclaimed it distinctly abnormal.


It is of interest to note that in the case of cheese j, gas was obtained
from a hole the size of one's fist, and that this contained practically
no hydrogen. The appearance of this hole was that of a strictly
normal eye except in size.
It was hoped that the gas of a typical "blow hole" could be ob-
tained. For this purpose a cheese containing such a hole was pur- ':
chased in Wisconsin. When it arrived at the laboratory it was found l
that the cheesemaker had punctured it.
From the results obtained it is clear that there are at least two
distinct types of gas formation.a The one is highly detrimental, and
is accompanied with hydrogen; the other is that demanded in a good
Emmental cheese. One is dependent upon the presence of sugar;
the other occurs in the absence of sugar. :
The presence of hydrogen in considerable quantities in the gas iso-
lated from Edam cheese by Boekhout and Ott deo Vries is very sug-
gestive of a gaseous fermentation of sugar, and to this Jensen has
ascribed the formation of gas holes in Edam cheese.
At this point it may be well to call attention to a source of error
overlooked by various investigators in their attempts to establish
the cause of any particular gas formation in cheese. Frequent exam-
ples are to be found in which gas production by bacteria in milk is
interpreted to mean that these bacteria can produce gas in cheese.
Although this may frequently be true, it must nevertheless be remem-
bered that the two media differ not only in chemical constitution but
also vary greatly in physical chemical condition.
Baumann,3 for instance, attributed the formation of eyes in hard
cheeses to Bacillus diatrypeticus case. From an experiment in
which this bacillus produced in milk gas containing 63 per cent of
carbon dioxid and the remainder almost entirely hydrogen, Baumann
concluded that the gas of normal as well as faulty eyes is carbon i
dioxid and hydrogen. The error of attributing the reactions of a
bacillus when cultivated in milk, which contains sugar, to cheese,
which after the initial fermentation contains no sugar, is so evident,
and the error in stating that the gas of normal eyes contains hydrogen,
without having first analyzed this gas, is so evident, that Baumann's
conclusions might be left unnoticed at this late date were they not
typical of several found in the more recent literature.
In all the analyses no appreciable amount of oxygen was found. |
The presence of large percentages of nitrogen with this absence of :
oxygen raises the question, Does air diffuse into the cheese with ab-
sorption of oxygen? Evidence of an active absorption of oxygen was
a This does not preclude there being a number of distinct fermentations or reactions of either type.

j :A



accidentally obtained. In attempting to study the gases produced
in sealed tubes a faulty form of tube was first used, which evidently
leaked. On attempting to exhaust, tih l)ows4t pressure which could
be obtained was 3.6 mm. It was soon ascertainedl that there was no
leak in the pump, but a leak in the tube was suspected. The tube
was left connected with the pump (connecting stopcock closely) over
night. The next morning 37.20 c. c. of gas was pumped out. The first
portion of 19.15 c. c. gave 4.57 c. c. of carbon dioxide and 2.21 c. c. of
oxygen. The residue was lost but was considered to be nit rogen. T1e
second portion was then pumped out, and of the 18.05 e. c. thus col-
S elected there were 4.85 c. c. of carbon dioxide, 1.45 c. c. of oxygen, and
the residue entirely nitrogen. The total oxygen amounted to 3.66 c,. c.,
S which, had it come by leakage, would have indicated an entrance of
= 13.7 c. c. of nitrogen. There was actually found 24.12 c. c. of nitrogen.
This leaves 10.42 c. c. of nitrogen to be accounted for. The carbon
S dioxid amounted to only 9.42 c. c. and, since the ratio 10.42 is much
{ larger than that obtained in other similar pumping where no leak
S occurred, it was suspected that oxygen had been absorbed.
To definitely determine this the apparatus shown in figure 3 was
used. With plugs of cotton at B, C, and in the bend above G, the
I tube was sterilized at 170 C. Then 28.5 grams from one of the Wis-
consin cheeses were carefully taken with trier and spatula flamed to
II prevent contamination as far as possible, and the plugs introduced
|E into A and sealed in as previously described. Mercury was drawn up
| into the tube E until it had just passed the stopcock F. After at-
I tacluhment had been made to the pump the whole was evacuated 5
iii hours and finally at a pressure of 1.2 mm. the capillary at H was sealed
ii off in a blowpipe flame. There was introduced into E 7.47 c. c. N.
il T. P. of oxygen from a tank. At the same time a sample of the same
il gas was taken for analysis, and found to contain 98.1 per cent of
:1 oxygen. Upon opening the cock F atmospheric pressure forced the
;| ~ gas over into the tube A. The mercury behind this gas was allowed
to rise until it had entered the capillary G. As close to this mercury
S as was possible G was then fused off with a blowpipe. There was
| left of the 7.47 c. c. introduced only a small bubble in the capillary,
and this at reduced pressure. After 6 days at 25 C. the tube was
opened by the usual method and the gas was pumped out and ana-
lyzed, with the following result:
C. c.
Total volume of gas collected.................................. 11.90
Carbon dioxid ................................................ 10. 96
Oxygen ...............................................------------------------------------------------...... 53
Residue, all nitrogen ....................................... .. .41
From the percentage composition of the 7.47 c. c. of gas added
at the beginning of the experiment it is known that 7.33 c. c. of




oxygen was added. At the end of the experiment there remained =i
only 0.53 c. c. of oxygen. There must, therefore, have been 6.80 c. . "I
of oxygen absorbed, or 0.239 c. c. per gram of cheese.
Such an active absorption of oxygen lends itself to the argument
that the nitrogen of the eyes found its way there by the diffusion in of
air. But, before such an argument can be con-
sidered valid, the following points must be deter-
mined: First, to what extent is cheese permeable
1 to gases in general and nitrogen in particular ?
Second, how much of the nitrogen present is due
to nitrogen dissolved in the cheese at the time of
Manufacture? Third, what evidences are there
S to show that the nitrogen does not arise in situ
from bacterial or chemical reactions?
Sn After various unsuccessful efforts to make an
F impermeable adhesive that would stick to
cheese, and so enable a slice to be sealed into
a diffusion apparatus, the following device was
made (fig. 4):
At B a membrane of plaster of Paris was
formed whose strength was reenforced by a per-
forated brass plate not shown in the diagram.
This membrane was desiccated until its perme-
ability was high, that of transfusion.10
Most of the air was forced out of D through
the membrane and E by raising the mercury.
A carefully taken disk of cheese was then
placed on the plaster of Paris bed. It was
gently held there while it was completely cov-
ered with mercury. Then, by lowering F, the
space in D was left under greatly reduced pres-
sure. This caused such a difference in pressure
between the upper and lower surfaces of the
FIG. 4.-Device for ascertain- cheese that the disk was held firmly against the
ing permeability of cheese to plaster bed, and the surrounding mercury was
gases' unable to float it. Preliminary experiments
showed that no mercury crept between the disk and the plaster,
and that the plaster did not become clogged with mercury or cheese.
After partial vacuum had been produced in D a few moments elapsed
before the gas retained in the plaster came to equilibrium. When
this was reached the mercury was carefully withdrawn from the top
of the disk of cheese until the surface was exposed. The mercury



left at the side prevented entrance of gas there, so that the only
path between the chambers A and D1) by which gas could enter D was
through the cheese.
The disks of cheese used were about 1 cm. in diameter and 2 to
2.5 mm. thick. They were taken from sound portions of freshly cut
cheese by means of a cork borer, and carefully sectioned with a sharp
razor. Every precaution was used in cutting and handling to pre-
S vent distortion and breaking of the texture. In one case the exposed
surface was that of an eye. The gas whose diffusion it was desired to
study was flooded into the chamber A. With both air and carbon
dioxid there was apparently no diffusion during an hour, even though
the pressure in D was reduced as much as possible. Longer experi-
ments were not practicable, because a continuous watch had to be
kept to see that no bubble of air entered through the rubber connecting
tube between D and F and altered the volume in D. With a trap to
prevent such a source of error the same impermeability for air was
observed during an experiment lasting several days.
Tis result was so remarkable that it was tested further in the
following manner: Instead of the parts E, D, F (fig. 4), a glass tube
led from B to a mercury pump. With the cheese slab C covered with
mercury the pump was operated till the lowest vacuum which could
be obtained was reached. By reason of the gas being given off by the
cheese, this was of course not so high a vacuum as the pump can
produce. When the vacuum was considered sufficient the pump was
allowed to rest in order to discover leakage, and, if there were none,
to allow the residual gas to distribute itself so that a reliable reading
on the McLeod gauge could be made. Then the mercury was care-
fully withdrawn from the top of the cheese, leaving its upper surface
exposed. Entrance of gas could now be detected by the McLeod
gauge. An experiment is given in detail below:
[Disk of cheese 7mm. diameter, 2.5 mm. thick, taken 10.50 a. m., Dec. 19,1911,15 mm. from the nearest rind.]

Timeof Pres-
reading.' sure.

a. min. Mm.
Apparatus exhausted, and, with cheese covered with mercury, pump pressure at......... 11. 14 0.075
Pum p resting....................................... ................................... 11.25 .140
D o................................................................................ 11.30 .150
Increase in pressure assumed to be due to gas evolved from cheese.
After 7 minutes pumping............................................................... 11.37 .070
Cheese exposed to air.................................................................... 11.40 .070
p. Tm.
D o................................................................................. 12.15 .150
Do................................................................................. 1.15 .270
Do....................................... .......................................... 1.50 .320
Do.............................. ... ........ .. ................................... . 2.15 .350
After 15 minutes pumping............................................................... 2.30 025
Cheese exposed to CO2 .................................... .. ........................... 3. 10 .060
Do ............... ..................... ............................................. 3.30 .090
After 5 minutes pumping............................................................... 3.35 .025
Cheese exposed to H7 ................................................................... 4. 1 .050
D o .................................................................................. 4. 30 .065
After 15 minutes pumping............................................................... 4. 45 .010
a. m.
Cheese left overnight exposed to air..................................................... 9.30 .430



When the disk of cheese and the mercury were removed air entered i
rapidly, showing that the plaster had not become plugged. Further- .
more, there was no evidence of mercury having crept between the 1
cheese and the plaster. It is not claimed that all the above listed
readings on the McLeod are very accurate, since the readings were i
sometimes made before equilibrium was obtained. All that was .
desired was the order of magnitude. Since the variation in tempera- .
ture during the experiment was only between the extremes 17 C.
and 19 C. and since the volume of the pump, gauge, and diffusion
apparatus was found to be 159 c. c., we may calculate from pressures
the approximate amount of gas which had apparently diffused through
the cheese. This amounted to about 0.09 c. c. during the first 5 hours !
and 0.09 c. c. during the last 17 hours.
Allowing nothing for possible small leaks, which were difficult to
avoid in the delicate manipulations required, the observed volume of
gas indicates a very remarkable impermeability. Practically the
same result was obtained with a disk of Cheddar cheese and other
samples of Swiss cheese.
The question at once arises, How to explain the evolution of carbon
dioxid which there is every reason to suppose does diffuse from cheese I
Van Slyke and Hart 17 found that a normal Cheddar cheese evolved
during 32 weeks 15.099 grams of carbon dioxid. Since they took
care to exclude surface growths of molds, it seems highly improbable
that this amount of carbon dioxid could have come to any great
extent from the surface layers alone. It must have diffused from
the interior of the cheese into the surrounding bell jar.
The following explanation will doubtless be found reasonable:
Becquerel 4 found that when the tegument of peas was mounted at
the end of a barometer tube, and a partial vacuum of 5 to 10 mm.
obtained upon the one side, with atmospheric pressure on the other,
thd tegument was impermeable to gas when dry, although permeable
when moist. In so far as the tegument of peas and a disk of cheese
are both Colloidal they may be compared. In the present experi-
ments the disks of cheese dried considerably both from exposure to
gases of low vapor content on the one side and the moisture free
vacuum on the other. By analogy with Becquerel's experiments
one would expect to find the dry cheese more or less impermeable.
Reference to the experiment detailed on page 21 will indeed show that
the permeability decreased as the time of the experiment increased,
or, in other terms, as the cheese became drier. Furthermore, in
an experiment in which the exposed surface of the cheese was kept
exposed to carbon dioxid, which was saturated with vapor, 1.04 c. c.
of gas was found to have passed through in 5 hours; ten times as
much as in the experiment with drying cheese.




It therefore seems probable that the permeability of cheese to gases
is due to the diffusion of dissolved gases, and that as the free solvent
becomes more and more attenuated the gas is more and more unable
to find its way through the gel.
Since in Emminental cheese a more or less dry rindl is produced, it
sems probable that little air can (litfuse into) the cheese. And from
the fact that in the manufacture of Cheddar a less dry rind as well as
a more open texture is produced, it seems probable that escape of
carbon dioxide more easily occurs in this type than in the Swiss type
of cheese.
S It must be remembered, however, that thle above experiments only
cover a very limited time, and that, even were the permeability as
low as the experiments seem to show, there is still the possibility that
S nitrogen may make its way slowly through the gel during the long
period of ripening. Possibly more extensive investigation would
S reveal that the larger percentages of nitrogen found in the eyes of
some cheeses are proportional to the age of the cheeses. Neverthe-
S less this penetration can only take place slowly.
The fact that penetration of air is so slow, together with the
avidity with which oxygen is absorbed, only tends to emphasize the
completeness of the anaerobiosis in the interior of the cheese, a con-
dition which Troili-Peterson16 found necessary for the best develop-
ment of the propionic bacteria.
| These experiments on the permeability of cheese to gases make it
evident that in pumping the gases from plugs of cheese we should ex-
pect the gas to be slowly evolved. Such was, indeed, found to be the
case. The reason for this was not fully appreciated at the time the
pumping were made, and it is very doubtful if all the occluded gas
4 was completely exhausted even after 20 hours exposure to high
vacuum. Reference to the experiment with plugs of cheese kept 6
days'in vacuo (p. 15) reveals the interesting fact that the amount of
gas evolved per gram of cheese was dependent more upon the state
of the vacuum than upon time. This is illustrated in the following
statement, in which the figures represent cubic centimeters of gas
evolved per gram of cheese per hour:

1 2
First 18 hoas .......................................................................... 0.0087 0.0042
Succeeding 6 days............................. .......................................... 0015 .0014
Last 18 hours .................................................. ........................ .0033 .0066

During the middle period, of course, the tubes were sealed, and the
S evolved gas increased the pressure. Evidently, then, the higher the
vacuum to which the sample was subjected the more rapidly was the
gas evolved, indicating that a considerable proportion of the gas was




dissolved or occluded gas rather than that formed during the time '
of the experiment. I
It may also be true that there is loose combination of carbon i
dioxid with inorganic salts, or with calcium and amino bodies, as in
the carbo-amino reaction, and that the stability of these compounds
is a function of the imposed pressure.
Let us now consider how much of the
nitrogen found in the eyes is attributable
to nitrogen occluded in the original curd.
One would expect the curd to be well
Saerated by the vigorous stirring it gets
during the process of manufacture. Mar-
J? \shall 14 has shown that aerated milk con-
S MI/AWRY" tains considerable quantities of nitrogen,
but, unfortunately for the purposes de-
sired, his data are only expressed in per-
centage composition and not very defi-
nitely in cubic centimeters of gas per
cubic centimeter of milk.
A A rough approximation of the amount
of nitrogen occluded in the curd was
GzLA-S obtained in the following way: A liter
< of milk was treated as in the process of
making Swiss cheese. When the curd
had reached the stage when it was suit-
able to hoop, the greater part of the
whey was decanted, and then the re-
sidual whey and curd were poured care-
fully into the glass cylinder A, figure 5
(inverted). As the curd settled, the
overlying whey was drawn off and more
of the mixture poured in. This was re-
peated until the tube was filled with curd
grains completely surrounded by whey.
The rubber stopper was then forced in.
SThe tube was next inverted to the posi-
FIG. 5.-Apparatus for determining tion shown in the figure, and the mer-
amount of nitrogen In curd. 0 '.
amount of nitrogen In curd. cury seal stopcock B was opened to re-
lieve the pressure. The rubber stopper was then forced farther in,
and the whey displaced by it escaped into C. By covering the
stoppered end of the tube with rubber-rosin cement and keeping
it under mercury, it was made perfectly gas tight. The cock B was
then closed, and, after the surplus whey in C had been drained out,
the apparatus was connected to the vacuum pump in the usual way.



When the pump and chamber C were completely exhausted, the
cock B was opened. It was found that the gas expanding in A
drove the whey almost completely up through the interstices of the
curd and into C.
An interesting point was observed. Comparatively little of the
gas came from the whey, while the major portion came from the curd
particles. Since a separation of whey and curd was accomplished,
it could not have been true that the gas evolved from the curd par-
tidcles originated in the whey, using curd particles as nuclei for the
formation of bubbles. Furthermore, there was comparatively little
frothing of the whey in C, most of the gas collected having bubbled
through C from A. Examination of curd particles will show why
this is so; for they have adhering to them minute bubbles, apparently
froth taken up during the stirring. It is quite evident that the col-
umn of whey in C through which the gas had to make its way pre-
vented a very complete exhaustion. Since the pumping was con-
tinued several hours and the tube then allowed to stand overnight
before the final pumping, this error was reduced to some extent. If
occasion arises to repeat these experiments this error will be avoided.
By the method described, 1.35 c. c. of gas was collected in one
instance and 0.86 c. c. in another. Of this, there was 0.58 c. c. of
nitrogen in one case and 0.39 c. c. in the other; average, 0.5 c. c.
The curd was roughly estimated to represent 20 grams of cheese.
Consequently there would be approximately 2.5 c. c. of nitrogen
per 100 grams of cheese. How this nitrogen would partition itself
between the body of the cheese and an eye is a question whose solu-
tion would be mere guesswork without further data.
While the 2.5 c. c. per 100 grams of cheese is a mere approximation,
and a figure which would vary not only with the extent to which the
curd is stirred, but also with the form of the curd particles and their
ability to absorb foam, nevertheless it is sufficiently accurate to
show that a large part of the free nitrogen found in cheese comes from
occluded air.

The question of whether any of the nitrogen found in the eyes is
set free in situ is a difficult one to answer, and one which can not be
definitely answered without further research. From the following
considerations, however, it is highly probable that it is not produced
during the course of that reaction which furnishes the gas to distend
the eyes. In those experiments in which samples from a cheese at
the period of its maximum eye formation were held in vacuo, the
nitrogen in the evolved gas steadily and rapidly declined in per-
centage, finally reaching almost nothing. This indicates that the
nitrogen collected was simply that dissolved in the cheese, and as



this was removed there was no evolution of free nitrogen to take iftii
place, such as occurred in the case of the carbon dioxid. '1]

.. *4i
The results of the whole investigation show clearly that the only -'
gas which plays an important r6le in the formation of normal eyes i
carbon dioxid. This is in entire harmony with the assumption which
has heretofore been accepted as a fact by various authors.
It remains to be seen whether there is any quantitative relation ||
between the amount of carbon dioxid evolved and that called for by *J
the process to which Von Freudenreich and Jensen ascribe the forma-
tion of eyes.
A study of the volatile fatty acids of Emmental cheese by Jensen 12 .
disclosed the fact that they are chiefly propionic and acetic, and that
often the ratio of these approximates 2:1.
Fitz 7 had previously shown that certain bacteria are capable of
producing this ratio of propionic and acetic acids-from lactic acid,
and he ascribed to their action the equation:
3C3H03=2C3H60+C2H402+C02+H20 "
lactic propionic acetic
Subsequently Von Freudenreich and Jensen8 isolated from Emmen-
tal cheese an organism which did ferment lactates according to the
above equation of Fitz, and whose introduction into cheese was fol-
lowed by an eye formation of which it was thought to be the cause.
The conclusion seems evident that here is an organism to whose
action may be attributed the formation of normal eyes.
The evidence is undoubtedly the clearest that has yet been presented.
There are, however, one or two points which will bear inspection
before the theory can be accepted as a full explanation.
According to the equation of Fitz three'molecules of volatile fatty
acids are accompanied by the liberation of one molecule of carbon
dioxid. Consequently it can be shown that a titer of 1 c. c. of
tenth-normal alkali for these volatile fatty acids should indicate the
liberation of 0.74 c. c. of carbon dioxid (N. T. P.). If, then, it is
found that the volatile acids from 100 grams of cheese neutralize 100
c. c. of tenth-normal alkali, and it is assumed that these acids are
acetic and propionic in the ratio in which they occur in Fitz's equa-
tion, we would have liberated 74 c. c. of carbon dioxid per 100 grams
of cheese.
This amount of gas is considerably more than is required to fill the
eyes, but the question remains how much is to be found in the body
of the cheese itself.
Reference to the experiments described on page 15 shows that at
an age of 55 days 37.2 c. c. of carbon dioxid per 100 grams of cheese



il were collected after the cheese has been held in vacuo one week. At
i the time of the experiment it was thought that this gas was produced
During that week. After the study wlich shows how i impermeable
Cheese is, this view had to be modified, for, even after considerable
i pumping, an appreciable quantity of gas must lhavc remained and
Appeared as "evolved"l' gas at thle end of the week. In order to make
a better estimation of the dissolved gas, plugs of this same cheese (at
San age of 4 months) were sliced into thin dislks to facilitate tlhe removal
Sof dissolved gas, and introduced into a tube. They were sealed in
J with the usual rubber stopper and rubber-rosin cement, and the tube
Joined to the mercti'y pump. After evacuating the pump the con-
Snecting cock was opened and the disks of cheese evacuated. The air
surrounding then in the tube was of course pumped out too. The
total gas thus collected after 5 hours continuous pumping contained
17.05 c. c. of carbon dioxide. The weight of cheese was 42 grams.
Hence, there were collected 40.6 c. c. of carbon dioxide per 100 grams of
i cheese (19 hours later 0.9 c. c. of carbon dioxid was collected, or 2.1
c. c. per 100 grams of cheese).
1 A duplicate determination gave 46.2 c. c. of carbon dioxid per 100
i: grams of cheese (with an additional 1.03 c. c. per 100 grams after 19
j hours). The average for the first 5 hours' pumping was 43.4 c. c. of
i carbon dioxid per 100 grams of cheese, and this we may fairly con-
S sider the quantity occluded at the time the plugs were taken. At
the same age (4 months) the volatile fatty acids corresponded to 40.9
c. c. of tenth-normal alkali per 100 grams of cheese.
Similarly, duplicate determinations of dissolved carbon dioxid and
volatile acids in an excellent imported cheese (No. i) gave the fol-
lowing data: Carbon dioxid per 100 grams, 67.8 c. c. and 54.8 c. c.,
average 61.3 c. c. Total volatile fatty acids in cubic centimeters of
tenth-normal alkali per 100 grams 95.1 and 97.7, average 96.4.
Assuming that all the volatile fatty acids were produced in strict
accordance with the equation of Fitz, the amount of these acids in
the first cheese indicates that there had been liberated 30.6 c. c. of
carbon dioxid against 43.4 c. c. found occluded; and in the second
cheese the liberation of 71.3 c. c. of carbon dioxid against 61.3 c. c.
S found occluded. There is a somewhat striking apparent relation-
S ship in this data, and the averages, 51.0 c. c. calculated, against
52.3 c. c. found, are in such close agreement that they are tempting.
A little consideration will show, however, that this agreement may be
S only accidental. At the time these analyses were made each of the
Cheeses had probably reached a state of little activity. The volatile
acids represent almost entirely the total amount produced in the
interior from which the samples for analyses were taken; while, if
we are to accept the results on Cheddar cheese by Van Slyke and Hart
as at all applicable to Emmental, it is certain that a considerable



quantity of carbon dioxid must have escaped in the months ia" i3
manufacture. Furthermore, although the actual volume of the eyes:
represents but a small portion of the gas in a given volume of cheese,
the normal volume of this gas in the eyes leaps into considerablA
significance when it is remembered that it must have been under ..
considerable pressure. That it is under pressure was made evident '
in some cases by its vigorous escape when using the puncturing
apparatus for its collection.
Unfortunately long delay in obtaining apparatus suitable for a
study of the gas escaping from cheese, as was done by Van Slyke and
Hart for Cheddar, have made it impossible to present any data on
this point. As before mentioned, the data on carbon dioxid evolved
from plugs of cheeses taken at the height of their gaseous fermenta-
tion and kept in vacuo a week is complicated by the fact that there
was probably a slow yielding of dissolved gas from the solid plugs as
well as the normal production of gas. Two other experiments, how-
ever, indicate to what extent carbon dioxid was being formed during
this period of maximum fermentation.
Portions of cheese W 2 from regions without eyes were carefully
selected and sealed up in a tube as described on page 14. The eye
membranes were carefully removed from a large number of eyes
and similarly treated. The tubes were simultaneously exhausted
with a Boltwood pump for several hours. Since in these cases the
cheese was in a more finely divided state, it is reasonable to assume
that predissolved gas was pretty thoroughly removed. After exhaus-
tion, the tubes were sealed off in a blowpipe flame and held at 25 C.
for seven days. At the end of this period the gas was collected:
34 grams eye membranes gave 14.95 c. c. of gas, 99.3 per cent of carbon dioxid,
or 44 c. c. per 100 grams.
36 grams from regions without eyes gave 10.06 c. c.. of gas, 98.2 per cent of carbon
dioxid, or 28 c. c. per 100 grams.
From this one pair of experiments it is not advisable to claim con-
fidently that the eye surfaces always produce the much larger quantity
of carbon dioxid, although this is plainly evident in the above case.
The significant fact is that such a large quantity was produced by
each region in the period of only one week. Of course it may be
claimed that although the division of the cheese was done in a dust-
free room and with sterile instruments, and the cheese introduced
into sterile tubes, yet the long manipulation admitted a heavy
reinoculation by bacteria, and that these produced a renewed evolu-
tion of carbon dioxid. Such an argument can not be completely
refuted, 'but the probability of a heavy enough infection is small.
The most likely source of carbon dioxid producing infection was by
molds, but these could not have grown in the complete anaerobic
condition in which the cheese quickly found itself.


S The following experiment serves to confirm tholast. Into a steri-
i Hzed combustion tube were quickly slipped pluhi of cheese taken
with sterile instruments. Each end of the tube was guarded with
cotton plugs. Carbon dioxid free air was then passed through, and
the carbon dioxide evolved from the cheese absorbed in the customary
train with all due precautions for exact estimation of carbon dioxide.
In the case of this experiment we would expect a higher amount of
Carbon dioxide, since there would be collected not only the carbon
dioxid produced, but a large portion of the predissolved carbon
Sdioxid. Such was found to be the case.
: One hundred and five grams in plugs taken from cheese WIV 1 when
| at the height of its fermentation gave:
SFirst 24 hours, 81.6 c. c. of carbon dioxid.
Second 24 hours, 66.7 c. c. of carbon dioxid.
Third 24 hours, 44.5 c. c. of carbon dioxid.
Fourth 24 hours, 63.7 c. c. of carbon dioxid.
S The increase on the fourth day was thought to be due possibly to
Growth of molds with which Van Slyke and Hart found difficulty in
their work on Cheddar cheese. The experiment was therefore dis-
Scontinued, although no growth was visible.
S A final word must be urged against the too liberal use of the equa-
tion of Fitz. As a terse representation of the probable relation of
the end products the equation has a legitimate use. As a compre-
hensive portraiture it is colored with presumption. The literature of
S fermentation is littered with equations, two or three members of
which are known to stand in certain quantitative relationships, while
the other members are given values which fit. This stoichiometrical
adjusting is particularly true of the gaseous products. One has only
to review the literature on the gas production of B. coli to assure
himself of the fact.
|| While Von Freudenreich and Jensen's use of the Fitz equation has
YI| been interpreted quite rigidly in the preceding pages, this was done
| simply as a test. From this basis alone one can not reasonably jump
|S to a final conclusion; but it must be remembered that the liberal use
i made of the Fitz equation was generous to the theory of Von Freuden-
S reich and Jensen in that all the volatile fatty acids, as determined by
: Jensen's method, were assumed to have been produced in accordance
S with this equation.
From a comprehensive view of the matter it appears to be quite
evident that the theory of Von Freudenreich and Jensen is not
capable of accounting for all the carbon dioxid produced. Indeed,
i t is not necessary nor expected that it should, but we have reached a
S point where it has become advisable to distinguish between a primary
and a secondary cause of eye formation, and to at least define clearly
what we mean when we attribute to any organism or to any reaction
the function of forming eyes.



Suppose that th5 propionic bacteria are active, but that they a-J
never sufficiently localized to concentrate carbon dioxid rapidly enou
at one point to produce an eye. In this case the gas would be mte"
or less evenly produced throughout the body of the cheese. N.w11
this state of more or less complete saturation of the body with cart.
bon dioxid is exactly the condition necessary for the most advaln ..E!"
tageous eye formation by any other reaction which may follow, el :-i.:
the gas evolved at a point would be largely absorbed and its inflating
energy dissipated. Of course it can be said that this saturation :
proceeds from the point where the eye is formed, and that the delay.-.:i
observed before an eye commences to grow represents the time
necessary to effect this saturation. :ii
This, however, is merely presenting the other horn of a dilemma
from which escape is possible only when the localization of the propi- '
onic bacteria is conclusively demonstrated. Gorini9 has contended j
that the localization of colonies may often be of as great importance
as their isolation; and it is interesting to note that he found no cor- .
relation between the colonies which stained on his sections of Grana i
cheese and the gas bubbles.
If, then, we distinguish between a "saturating" gas production
and an "inflating" gas production, we will have at least defined a !
possibility which must be squarely met, and a hypothesis which may |
lead to a differentiation between a primary and a secondary cause of
eye growth. ,
The favorable results obtained with cheese inoculated with pro-
pionic bacteria indicate that they may play an important r6le. But
is this r6le primary or secondary? Is it a strictly localized action or
is it simply the provision of that saturation without which some
primary and strictly localized reaction would be without avail? The '
same question arises in the case of the glycerin-fermenting bacteria >:4
to which Troili-Petersson 16 has ascribed an important r6le in the :
holing of Swedish cheeses. In fact experimental cheesemaking of the i
past, though not so thoroughly controlled as in the experiments of
Troili-Petersson and those of Von Freudenreich and Jensen, bear:.
evidence that any one of a number of gas-producing bacteria nmayr
provide the saturation, not to mention those reactions which Van :
Slyke and Hart 17 have proposed as contributing to the carbon dioxid :i
in Cheddar cheese. On the other hand, any one of these may be the
primary "inflator" and the other the secondary saturatorr." .
In this connection it may be of interest to note a peculiar phe- i
nomenon met with in some experimental cheeses. A number of -J
these made with artificial" rennet by Mr. Doane were reported in
their early stages to have begun a normal eye formation. Seldoa, *
however, did this beginning develop into a normal holing. These
cheeses were of small size, and, since it is known that small-sized
cheeses for some reason not yet clearly defined seldom develop large :
..... :!ii. .



eyes, the failure in these cases may on general principles be vaguely
attributed to size. However that may be, it was found upon pump-
jng out the dissolved gas that their amount was low. "1he three
cheeses examined were 39-45, 39-11-2, and 46-4-1. (See Table 2,
p. 12.)
It is well known from the work of Jensen and others thliat the bac-
teria found in "natural" rennet are often distinct from those found
in "artificial" rennet. Since the cheeses under discussion were made
with the latter, is it not possible that the reaction which started( the
eye formation was rendered inadequate because the gas-producing
propionic bacteria, which might have saturated tlihe cheese with
carbon dioxid, were absent? That the observed holes were truly
the beginnings of normal eyes, and were not a pinhole formation
resulting from an initial gaseous fermentation of sugar, is evinced
by the fact that hydrogen was absent.
Exhaustive research alone can unravel this tangle; but it is hoped
that the present investigation has provided both a clearer definition
of the problem and a sound basis of fact.
1. The gases of normal "eyes" in Emmental cheese are exclu-
sively carbon dioxid and nitrogen, and of these only the carbon
dioxid is of significance.
2. The nitrogen accompanying the carbon dioxid in normal eyes is
that of air originally occluded in the curd at the time of manufacture.
3. There sometimes occurs during the initial fermentation an evo-
lution of gas characterized by the presence of hydrogen. This is
believed to be due to the gaseous fermentation of sugar.
4. The hydrogen from such an initial fermentation may sometimes
linger to contaminate the gas of normal eyes.
5. The two fermentations are distinct and are characterized by their
gaseous products. The one is detrimental, the other that demanded
of a good Emmental cheese.
6. High oxygen-absorbing power combined with low permeability
of the cheese to air render the interior thoroughly anaerobic, and conse-
quently favorable to the growth of anaerobic bacteria.
7. A comparison between the amount of carbon dioxid evolved
and the total volatile fatty acids shows that the activity of the pro-
pionic bacteria of Von Freudenreich and Jensen is not sufficient to
account for all the carbon dioxid found.
8. It was found that cheese is capable of retaining a very large
amount of carbon dioxid.
9. The possibility is suggested that there are two phases in the for-
mation of normal eyes, a saturation of the body with carbon dioxid,
and an inflation of eyes; and the bearing of this hypothesis on the
production of gas by a specific cause is discussed.


. .. .......

1. [BXCHLER, C. Beitriage zur Erforschung des Giihrungsverlaufes in der Emo ki.
thaler Kiaefabrikation. Schweizerisches Landwirtschaftlichen Centza)!bl',
Heft 1-6,1896, cited by Jensen.]
2. BOEKHOUT, F. W. J., and OTT DE VRIES, JAN JACOB. Sur deux d6fautso
fromage d'Edam. Revue G6n6rale du Lait, vol. 8, No. 14, p. 313-322; .No-
15, p. 347-356. Liege, Sept. 30, 1910..
3. BAUMANN, FRITZ. Beitrage zur Erforschung der Kisereifung. Die Land.wi t: .i
schaftlichen Versuchs-Stationen, vol. 42, p. 181-214. Berlin, 1893. 1 I
4. BECQUEREL, PAUL. Sur la perm6abilit6 aux gaz de atmospheree, du t6guxaeniit
de certaines graines dess6ch6es. Acad6mie des Sciences, Comptes Rendmus :.
vol. 178, No. 22, p. 1347-1349. Paris, May 30, 1904. 1
5. DENNIS, L. M., and HOPKINS, C. G. Die Bestimmung von Kohlenoxyd, Methua ,
und Wasserstoff durch Verbrennung. Zeitschrift ffir Anorganische Chemie,
vol. 19, p. 179-193. Leipzig, 1898.
6. FISCHER, EMIL, and PENZOLDT, FRANZ. Ueber die Empfindlichkeit des Gerumc
sinnes. Annalen der Chemie, vol. 239, No. 1, p. 131-136. Leipzig, 1887.
7. FITz, A. Ueber Spaltpilzgihrungen. VI. Mittheilung. Berichte der Deut i-
schen Chemischen Gesellschaft. Vol. 13, p. 1309-1312. Berlin, 1880.
8. VoN FREUDENREICH, EDWARD, and JENSEN, ORLA. Recherches sur la ferment .
tion propionique dans le fromage d'Emmental. Annuaire Agricole de la Sui'e.,
vol. 7, No. 4, p. 221-242. Bern, 1906. I
9. GORINI, CONSTANTIN. Sur la distribution des bactdries dans le fromage de Grant. 4
Revue Gen6rale du Lait, vol. 3, No. 13, p. 289-293. Lierre, Apr. 15, 1904. :
10. GRAHAM, THOMAS. Chemical and physical researches. Edinburgh, 1876.::
11. JENSEN, ORLA. Studien fiber die Lochbildung in den Emmenthaler Kes... |
Centralblatt ffir.Bakteriologie, Parasitenkunde und Infektionskrankheiten. ,
Abteilung 2, vol. 4, No. 6, p. 217-222, Mar. 22; No. 7, p. 265-275, Apr. 1; No. :bi.
8, p. 325-331, Apr. 26. Jena, 1898. i
12. JENSEN, ORLA. Studien fiber die flfichtigen Fettsauren im Kise nebst Beit-
rigen zur Biologie der Kiisefermente. Centralblatt fur Bakteriologie, ParW
sitenkunde und Infektionskrankheiten, Abteilung 2, vol. 13, No. 5/7, p. 161-17-9. S
Oct. 7; No. 9/11, p. 291-306, Oct. 21; No. 13/14, p. 428-439, Nov. 1; No. 16/1, 'I
p. 514-527, Nov. 11; No. 19/20, p. 604-615, Nov. 26; No. 22/23, p. 687-705, ;
Dec. 10; No. 24, p. 753-765, Dec. 28. Jena, 1904. AN
13. [KLENZE. Handbuch fufr Kiisereitechnik, p. 198, cited by Jensen.]
14. MARSHALL, CHARLES E. The aeration of milk. Michigan Agricultural Experi- A'.::;
ment Station. Special bulletin 16, Agricultural College, 1902.,
15. NUTTING, P. G. The spectra of mixed gases. Astrophysical Journal, vol. 19, 1
No. 2, pp. 105-110. Chicago, Mar. 1904. !
16. TROILI-PETERSSON, GERDA. Experimentelle versuch Uiber die reifung muid
lochung des schwedischen giiterkses. Centralblatt fOr Bakteriologie, ParmR
sitenkunde und Infektionskrankheiten, Abteilung 2, vol. 24, No. 13/15, p.
343-360. Jena, Sept. 8, 1909. :
17. VAN SLYKE, Lucius LINCOLN, and HART, EDWIN BRET. The relation of carbon :
dioxide to proteolysis in the ripening of Cheddar cheese. New York Agricul-**W'0
tural Experiment Station (State) Bulletin 231. Geneva, 1903.

':: ]i
* i

j44 4+


L.^ vr ; '
: ..

W r.. ...
C.., ..:,,,,

I1%yE \ 1 .-.
,* ;.. *" ,* ,

' *-.,. ,4
, .. .. :


, ,' : ^ ...*.. -
1 . ...



r6.. ".
4 *i N." .
Vi' .. ".. ........ ... o .. ~ ,

"4 -* *
Ei -:" .: ":. -. '

t- i',,

.: ,,,

; 1,'I *

.. ..*. -,:. .....
S "** .,. ." ,:' i . *. : ""
:".< ^ *-, .... : ,. *
:'""~ ... .' :: .:" :". ." ;% -'.
.." 'H
^ *:..- :; ,- * -
i! J '.'.^ : _... ; : ** . ..


-! .... '
"". :-* "; ..


. .f

* 4

... E I:


: ..


-O *WN
.::** "' *" ^ :

. ....*:--t ..^ t .

*. . -. .

:; ? :::'-::i

4 1.

. "r*' :^' .. l
: ..r,*."'


/ ::

W "






3 1262 08928 7311

..-WA 1 ~4 yj.4~. x Z._ 111&El KI E,1U~
", t "I." ,f w .. ..
Lrfi "W 1'"-T

.. ......

X'% ~' * .A',
4:. .A. '. ., . ".. "..M i, ..

.r2'4r m r' '~ :: ""j
Ye "I >.- ' '. ,.. .'t. ... A~'~%~~k~

.. .... .....4

1-.Z 'i q-1. k~Irrw 1r
.........-.. , ., .; .......

.A -44.-
t ia. ... { %. A.. '"

... ,, ,
7'f; .- ,.: ,,i..o,

.. . . . 0

Ape a .
'.la'! ;'.0"
....' "'.. .... ::.r-, .

421. 1F... ..
i {* r . .t .-, Q Q: .
4 .... < ," . .,:. .,.'#.S t jqC-
t" '- V", A '' .. ; I'T " -: '

C ..g"..4q,..rj" J *. "'."I%" "'. C' "
isi 4- '% ." - ,-.,. -..-, .,.

I- . . 4.
-'." .,. , ' -. . ''."V .;. .,. j._~4
... k 'T c,;. .. .,.;.q.")
0 ... t K . "4 . &,. ': ?',
:tv ."- kj 'i i;'. ""

...... t j ilk t.
p. ., .. t -.. "" ," .,s r* A :..K t t

Jr. J-F r.,
," A.:l " .. ', .1 i 1r,4 1

..2 *: .... .. t ....

Y,' 4',&t"
-.. 1 ,. '4 ,L 4
" "-'4r' "
,. my 7,, .,

-, .. .- '... f, 4 ...." .

M-. f' Z. A .
A .A,
r n ..1 :' ski A'= ". : ..: .'
'"" "7" ...1 .. . "':: ... 44 jJ 44 t
_j; P .'' IVY?

.. ... d

SW ON .'4
,.' : 4J. I.,

-7 - -- : 'z o:.
" .. 4 .
o A --

4!. .
." -," ['. r

4.4 p'. 4

~ -
4 i, '.a :. ,-. .-,, -" .. ,. *: '-, Y.

J-1" -4., 4

AN ,. EN