U. S. DEPARTMENT OF AGRICULTURE,
BUREAU OF ANIMAL INDUSTRY. Bulletin No. 71.
0. E. SALMON, D. V. M., Chief of Bureau.
CAMEMBERT TYPE OF SOFT CHEESE
THE UNITED STATES.
BY
IT. W. CONN, CHARLES THOM, A. W. BOSWORTH,
W. A. STOCKING, JR., AND T. W. ISSAJEFF.
GOV
WASHINGTON:
jr CAL!
IGELES
SEP 23 1952
LIBRARY
GOYT. PUBS. ROOM
] :
OK<; AM/ATION OP THE HI liKAl OF VMM A I. IMM STKV.
'. V. M.
A. D. MKI.VIN, D. V. S.
1. 1.. M.. M. D.
!'.!>. 11. \\ U.S., chid': Ci. \KK\CI: I',. I. AM. I'
chief.
Inxju'i-tlmi I)ii-'ix'nin: \. M. FAI;U\<. i"V 1'.. S., 1'. \'. M., chief.
(}<i<iriit'<n,' I>tris!i>n: IxiciiAKD \V. IlicKMAN, I'h. <.., \'. M. !>., chief.
AW-' :. TIK>\]I-.>N. M. S.
Art id: W. S. D. }\^
Animal Husbandman: QTSQRGE M. ROMMEL, B. S. A.
( )iu;i:i.'i .
i. \i;i; \T<
/>;//>;-.,,. M \HK.\- DORSET, ^1. !>., chief.
ri^iiin: .Fnii.N K. M < >n i.ici:. A. M.. V. M. !>., cliicf.
/',Io</ir<ll Dii-it'iu,,: r.UAVTdV II. 1, SC., A. M.. in
KXI'KKIMKNT .-TATIdN.
K. ( '. SCHIKH-UKI:. M. I >. V.: expert assistant, \\ . K. C'urrox.
i\-l I HAK',10.
l>r. M. o. Ainl-Tviii. care Qeo. A iioniu-i
in. Minn.
Dr. linn C. Aver. I'n-t-t Mlu-.- Buililinir.
Dr. F. I.. Do \Vnlf. curr Cha-. \VollT I'ai-ki-
Dr. F. \V. Ainsworth, Union stock Yards, I'itt.s-
To[>eka, K
Dr. Geo. Diiewif,'. ''are rnion Stock Yard
finnati. Ohio.
Dr. K. I' Do\vd. eare White, I'evcy A Dexter Co.,
ha. Nehr. \Von
Dr. 0. 8. Baker, 6th and Townc -.< Fran- Dr.' o. ):. Dyson, 316 Exchange Building, Union
. Yard-. Chicago. 111.
Dr. L. K. Baker. South St. .I..M- r h. Mo. Dr. Qeo. C. Faville. I'. ( ). l*)x 7'.'.. Norfolk, Va.
Dr. Boyd Baldwin, eare Cudahy Bros.. Cndaliy, Dr. .1. \V. Fink, care Swift it Co., Harrixm Sta-
\Vis. 'ion, Newark, N. .1.
Dr. A. K. Bchnkc. r....in !::_'. Federal Building, Dr. T. A
Milwaukee. \Vi-. land.
Dr. John A. Bell. \Vatert. .wii. N. Y. Dr. II. II. c, .hn-oii M.. Louisville, Ky.
Dr. S. K. Bennett, room :<::s. Live -t.,ck K\ Dr. \V. II. (iil.l.s. eare Morton
.ire Building. Kansas city. K hra^ka City. Nehr.
Dr. K. L. Bertram: . Dr. I., i-. -iiniond. Sta;
Iowa. fich.
Dr. Fred Braginton, care Continental Packing Co., Dr. II. A. lle.iriek. -Ji.'i st. I'aul st., Bait
BloomiiiKton. 111. Md.
Dr. J. J. Brougham, eare Mi- Yard-. Dr. <>. T
Mo. Mr. (i. S. lliekox, I', u. box !
\V. Butler, eare Druimnond Bro-., Kan I'l.-ili.
Clairi Dr. A. A. B . 111.
Dr. J. B. Clancy, Xatioiir Is, 111. Dr. \V. K. Iloue, eare \VrMern 1'aekiiiK '
Dr. Lowell Clark. ncy Buildin.
Denver, Colo. Dr. Julius li
Dr. Charles Cowie. Ogdensburff, N". Y. \. J.
Dr. Da vid Camming, 912 Lapeer are., -Port Huron. Dr. F. \v. Hontingti
Mich. K. K. uhari, Toitlaiid. Me.
Dr. Kobert Darlinu. care Chn.s. K. Hardy, San Dr. I;..;
Cal.
Vebr. Dr. G ii, KxehaiiKe BnilililiL-
Dr. J. F. Deadman, Sault Ste. Marie. Mich. Iowa.
Mr. Albert Dean, room >, Stwk Yard Station. Dr. James J<
Kansas City, KULIS. cbn]>el-! .ml.
.tided on third |>a_
U. S. DEPARTMENT OF AGRICULTURE,
BUREAU OF ANIMAL INDUSTRY. Bulletin No. 71.
0. E. SALMON, D. V. M., Chief of Bureau.
THE CAMEMBERT TYPE OF SOFT CHEESE IN
. THE UNITED STATES.
BY
H. W. CONN, CHARLES THOM, A. W. BOSWORTH,
W. A. STOCKING, JR., AND T. W. ISSAJEFF.
WASHINGTON:
GOVERNMENT PRINTING OFFICE.
1905.
LETTER OF TRANSMITTAL
U. S. DEPARTMENT OF AGRICULTURE,
BUREAU OF ANIMAL INDUSTRY,
Washington, D. C., March 23, 1905.
SIR: I have the honor to transmit herewith a manuscript entitled
' ; The Camembert type of soft cheese in the United States," and to
recommend its publication as a bulletin.
This paper represents the initiatory effort in a series of cooperative
experiments which has been undertaken by the Dairy Division of this
Bureau in conjunction with the Storrs (Conn.) Agricultural Experi-
ment Station. The full scope -and character of these experiments are
explained in the introductory chapter on the next page.
It is anticipated this work will result in much benefit to a large
class of our farmers and dairymen, who may thus find a new and
highly remunerative outlet for their surplus milk supply.
Respectful!}',
D. E. SALMON,
Chief of Bureau.
Hon. JAMES WILSON,
Secretary of Agriculture.
2
INTRODUCTION.
The investigations in soft-cheese making were undertaken by the
Storrs Agricultural Experiment Station with the hope .that there
might be developed a new industry especially adapted to New Eng-
land conditions. The Camembert type of soft cheese seems espe-
cially adapted to home manufacture. The process of manufacture
has, however, been held as a trade secret and no source of information
has been available for those who may have desired to make this type
of cheese. A series of preliminarj^ studies was undertaken, and the
results gave promise of a successful solution of the problems involved
and demonstrated the value of the work.
The continuation of the investigations, however, involved a larger
outlay than the means of the station would permit. Recognizing the
w r ide value which would result should the experiments prove success-
ful and a new industry be established, a cooperative agreement was
entered into between the Storrs Agricultural Experiment Station and
the Dairy Division of the Bureau of Animal Industry of the United
States Department of Agriculture, Washington, D. C. Under the
terms of this agreement a thorough study is to be made of the prob-
lems involved in the manufacture of some of the types of soft cheese
which are so popular in Europe, but which have l>een manufactured
in this country to a very limited extent only.
The Storrs Agricultural Experiment Station has furnished the
necessary laboratories, apparatus, and curing rooms for conducting
the experiments and has had the general planning of the work. The
Dairy Division of the Bureau of Animal Industry has furnished
three exjx i rts to assist in the work, namely, a mycologist, a chemist,
and a practical cheese maker, and has provided for certain traveling
excuses connected with the work. The organization of the actual
working staff connected with these investigations is as follows:
H. W. Conn, bacteriologist of the Storrs Station, sii|xrvisor.
Charles Thorn, cheese exjx>rt of the United States Department of
Agriculture, mycologist.
A. \V. Bosworth, cheese expert of the United States Department of
Agriculture, chemist.
Theodore IssajefT, expert of the United States Department of Agri-
culture, cheesemaker.
4 INTRODUCTION.
W. A. Stocking, jr., assistant bacteriologist of the Storrs Station.
In addition to the foregoing, who have been actively engaged in the
cheese investigations, credit is dne to Dr. B. B. Turner, chemist of the
Storrs Station and of the Connecticut Agricultural College, for val-
uable suggestions, and to Mr. W. M. Esten, laboratory assistant to
Doctor Conn, for bacteriological determinations. The Connecticut
Agricultural College has freely permitted the use of laboratories and
apparatus for this work, without which we would have been seriously
handicapped.
The work is now thoroughly organized and the investigations are
being conducted with the following objects in view :
(1) To learn the details of the methods of manufacturing some of
the soft cheeses.
(2) To learn more accurately the details of the process of ripen-
ing, and to determine the organic agents involved and their exact
functions.
(3) To study the chemical composition of the ripened product, the
changes involved in ripening, and their relation to food values and
digestibility.
(4) To discover the difficulties connected with the manufacture of
soft cheese and the ways of meeting them.
(5) To learn whether the manufacture of a first-class product is
possible in the United States.
(6) To make public property of all useful and desirable informa-
tion upon the subject of the manufacture of soft cheese.
This bulletin is necessarily preliminary in its nature, but it will be
followed soon by others dealing specifically with the problems in-
volved in soft-cheese making and how to solve them.
L. A. CLINTON,
Director Storrs Agricultural Experiment Station.
En. H. WEBSTER,
1 '
Chief of Dairy Division, Bureau of Animal Industry.
CONTENTS.
Paae.
Soft cheeses compared with hard cheeses 7
The market for soft cheeses in the United States 8
Present market conditions of Camembert cheese 12
The true Camembert type - --. 12
Reasons for selecting Camembert cheese 13
European literature consulted 13
Outline of our method of procedure 15
Ripening of Camembert cheese. 15
Compendium of our scientific investigations 17
Mycology - 17
Bacteriology. 1 22
Chemistry - 24
General conclusions 26
Control of th:> bacteria and molds 27
General summary . 28
Another report proposed . . 29
5
ILLUSTRATIONS.
Page.
PLATE 1. Fig. 1. A board of Cameiubert cheeses. Fig. 2. An improperly
ripened Camembert cheese ; the outer rim is liquified, while the
center is hard sour curd --- 1"
2. A series of Camembert cheeses. Fig. 1. Unripened cheese.
Fig. 2. Half-ripened cheese with well-developed moldy rind.
Fig. 3. Cheese two-thirds ripened, but with central mass of
hard curd showing some gas holes. Fig. 4. Cheese nearly
ripe and so soft that the upper edge moved because of its
weight during the exposure 17
6
THE CAMEMBERT TYPE OF SOFT CHEESE IN THE
UNITED STATES.
Introductory Paper.
By H. W. CONN, CHARLES THOM, A. W. BOSWORTH, W. A. STOCKING, Jr., and
T. W. ISSAJEFF.
SOFT CHEESES COMPARED WITH HARD CHEESES.
Hard cheeses have for a long time been well known in this country
and have been manufactured by our dairymen in large quantities.
Until recently little has been heard concerning another type of cheese
properly called soft cheese. This latter type, although made from
curdled milk, requires very different methods of manufacture, and
when ready for the table is a totally different product. As the name
indicates, the cheese is softer, frequently so soft as to be readily
spread upon bread like butter. But in addition to this there are
other striking differences between the soft and hard cheeses. The
soft cheeses are smaller than most of the hard cheeses. They do not
keep well. Above all, their flavor is quite different, being usually
much stronger. It is a well-recognized fact that while cheese is a
valuable food because of the large amount of protein it contains, nev-
ertheless it owes its chief popularity to its flavor. Flavors are known
to lx physiologically necessary to the proper digestion and assimila-
tion of foods. Hence the value of highly flavored foods that give
relish to those coarser and less flavored can hardly be overestimated.
The soft cheeses, with their strong flavors -so thoroughly enjoyed by
some people, have, therefore, a value decidedly their own.
The primary difference between the hard and soft cheeses is due to
different methods of manipulation of the curdled milk. In the hard
cheeses (Cheddar, Swiss, American, Edam, etc.) the milk is curdled
rather rapidly and the curdled mass subsequently cut into small pieces
to allow much of the whey to separate from the curd. Sometimes the
curdled mass is heated still further to separate curd and whey. After
this the curd is placed in frames and subjected to a high pressure (hat
forces out more of the whey, leaving a very hard mass of tolerably dry
curd. Such a curd ripens slowly and is not ready for market for some
months, and even when perfectly ripe it preserves its form and never
becomes very soft.
24424 -No. 7105 M 2 7
8 BUREAU OF ANIMAL INDUSTRY.
The soft cheeses Cameml>ert, Brie, Isigny, Limburger, Backstein,
Neufchatel, cream cheeses although each is made in a special man-
ner, all agree in one point, namely, the whey is never fully drained
from them. The curdled milk is commonly ladeled into forms and
allowed to drain naturally. Soft cheeses are not subjected to pressure
or heat, and therefore contain a larger per edit of water at the start
than the hard cheeses. As a consequence of their high water content
and soft texture, they afford favorable conditions for the growth of
various microorganisms, and en/yme action also occurs more readily
than in hard cheeses. The action of these two groups of fermenta-
tion agents ripens the cheeses rapidly and develops high flavors.
They are ready for the table in a few weeks instead of months, and
when ripened have not only strong flavors but frequently very pene-
trating and more or less offensive odors, as in the well-known Lim-
burger cheese. Because of the rapidity of the chemical changes going
on in them, these cheeses are more perishable than the true hard
cheeses. Their marketable period is very short. If placed upon the
table at just the right period they are very appetizing. If insuffi-
ciently ripened they do not have the proper flavor or texture, and if
even slightly overripened they rapidly show r a tendency to decay and
are soon ruined. To place them on the market in such a way that
they reach the consumers' table at the proper stage of ripening is one
of the difficult problems of the industry.
THE MARKET FOR SOFT CHEESES IN THE UNITED STATES.
The home of these cheeses is continental Europe. No part of
Europe is 'without its types of soft cheese. There are several gen-
eral types like Limburger, Camembert, and Brie that are widely
distributed and made in different countries, and there are, in addi-
tion, many local varieties with a narrower reputation. From their
centers of manufacture they are exported to various cheese-consuming
countries, so that all of the chief types are generally found through-
out Europe. Importations of soft cheeses to the United States have
not been very large. The American market in the past has been
almost monopolized by the hard cheeses, principally of the type of
the English Cheddar. Americans have known little of the great
variety of soft cheeses common in European markets, and then 1
has consequently been ver,y little call for them. Within the last
generation a moderate demand for soft cheeses has been developed.
This demand may be attributed partly to our increasing foreign
population and to the desire of people of European birth for the
types of cheese familiar to them at home, if is also partly due to
the taste acquired by Americans who have traveled abroad, as well
as to a growing appetite for them among others who have tried the
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 9
new delicacies. This has led to the importation of several varieties
of soft cheese to satisfy a market which is increasing every year.
This importation has reached about a million pounds, with a value
of about $150,000 per year.
There are many reasons why the total importations have not been
greater. It is difficult to import the more delicate varieties in prime
condition. Two weeks is the shortest period possible between leav-
ing the factory and reaching the retail market in America. If
fully and properly ripened before shipment these two weeks cover
almost the entire marketable period of the cheese. To reach our
market in condition for use at all they must be packed and shipped
before they are in condition for the table. In practice some are
shipped before they are half ripe or even when the ripening has
just begun. The change of conditions and entire lack of care in
transit makes the resulting cheeses which reach our markets of very
uncertain quality. It is also difficult for the casual buyer to deter-
mine .when the cheese is really ripe. These irregularities furnish
good reasons why such importation is discouraging to dealer and to
consumer alike.
On the other hand, these conditions have stimulated the production
of certain soft cheeses in America. Particularly is this true of Lim-
burger cheese, for which there is a large demand. It has been found
not difficult to manufacture, and it is now made in this country in a
number of places with a considerable degree of success. In response
to a similar demand a considerable business has arisen in the manu-
facture of the forms sold in our markets, as Brie, Isigny, Wiener,
lunch, miniature, etc., all of which, though sold under all sorts of
trade names, have a common type of ripening. Although the details
of form and making vary with the factory, the brand, and the season,
these may be grouped together perhaps as the "American Brie " type.
This American Brie type must not be confused with the French Brie
type, for the latter is quite a different product. These American
types have made for themselves a large market and may be regarded
as a fairly well-established product. A preliminary study of their
ripening process and the factories in which they are produced indi-
cates, however, a very unsatisfactory condition. The ripening is so
little understood and results are so uncertain that one maker admitted
that after the cheese is made the flavor of the product is practically
an accident over which he has no control. Some of the cheeses are
ruined in the ripening and the factories that make them lose consider-
able of their product. It is evident, therefore, that the manufac-
ture of the American tyjx 1 of these cheeses can hardly lx regarded
as a success until some means of controlling the ri]M>ning can IH>
devised that shall reduce the failures and give a more uniform
10 BUREAU OF ANIMAL INDUSTRY.
product. Moreover, the American type, as before stated, is quite dif-
ferent from the French; it is stronger, has a more unpleasant odor,
and a less delicate taste. The delicate tasting Brie and Cameinl>ert
cheeses are mostly imported from France. Up to the present time
their manufacture has been successful in but few places in America,
and there is a general belief that their manufacture in this country is
impossible.
Another indication of the growing popularity of soft cheeses is
the large demand for the varieties sold under various brands as
Netifchatel and cream cheese. The basis for most of these is the
common " Cottage " or " Dutch " cheese, and many of them are merely
sweet or sour curd put in an attractive and appetizing form. The
manufacture of these varieties is comparatively simple, with prac-
tically no danger of loss. They are sold and eaten when fresh.
Further, they will yield a larger amount of salable cheese from the
same quantity of milk than other varieties, and they bring prices
almost as high as the best imported cheese. Very naturally these
cheeses form the basis of a profitable industry. But here, too, we
find that the common American product is different from the cheese
in Kurope going under the same name. The Neufchatel cheese of
Europe is a ripened cheese, but the cheese going by this name in the
United States is usually unripened curd with an attractive wrapping.
Some of these American products, however, are highly flavored and
compete favorably with certain ripened types that are imported. All
of them command a far higher price than the ordinary hard cheese.
It is certain, therefore, that there is an established and rapidly
growing market for the best types of soft cheese. The profit in the
manufacture of the hard types of cheese is exceedingly small, both to
the factory and to the producer of milk, but the prices of the highest
class of soft cheeses (40 cents or more per pound) are so much higher
than the prices of the hard cheeses that the introduction of their
manufacture should be a great benefit to the dairymen. The fact
that soft cheese will not keep very long gives an especial advantage
to the domestic over the foreign producers if the products can be
made of equal merit. For such perishable products there is a great
advantage in an intimate relation between the maker and the con-
sumer. Factories near enough to a large city to supply the demand
with cheese always ripened to exactly the prime condition should
easily be able to control their home market if they can once control
their product. It seems possible that there will be an advantage,
nearer the small cities at least, in their production by individual
dairymen who could dispose of their own product in a restricted
market.
The great difficulty in the way of the successful development of
the soft-cheese industry is in the difficulty of controlling the ripening
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 11
so as to produce a uniform product. Even in the localities where
these cheeses are made in Europe their treatment seems to be purely
by " rule-of-thumb " methods, handed down as the result of more or
less successful experience, rather than by an accurate knowledge of
the reasons for the various steps in the process. As a natural COIKC-
quence, the results are to a large degree irregular, and utterly unex-
plained failures of many kinds are familiar occurrences. Recognizing
the difficulties in their manufacture in Europe, it is not strange that the
troubles increase when attempts are made to transplant such indus-
tries into a new country, where the practical rules applicable in the
old locality are vitiated by new sets of conditions, such as atmosphere,
temperature, flavors of milk, and other factors. So great have been
these troubles that there is a belief that the American product can
not be made equal to the imported. It is said that the difference in
the food which the cows eat produces a differently flavored milk,
thus making domestic cheeses unlike those imported. Whether this
be true or not, it is certain that little success has been reached by
American makers in trying to reproduce the higher, more delicate
grades of European soft cheese. When they have succeeded it has
been by importing their help from Europe, and in these cases the
makers have carefully preserved their methods as trade secrets. Even
in the successful factories many failures occur. Little attempt has
been made, apparently, to overcome the new difficulties arising in the
new conditions. As a result the home product has commanded an
inferior price and is less esteemed than the imported cheese. For
this reason the American makers have disguised their goods by French
labels and have not tried to make a market for an American product.
The market varieties show great irregularities. Complaint is con-
stantly made by thCj. dealer and consumer as well that common brands
of some of the soft cheeses are utterly unreliable. In the purchase
of a cheese, except by an expert, it is little more than a lottery
whether it will be mostly sour curd or might 1x3 eaten through a
straw. It is clear, then, that although there is a good market for
uniform and reliable goods, radical changes must l>e made in the
methods of production l>efore these soft cheeses will obtain the large
place in the American market which their real merit deserves. For
the American dairyman the development of this almost new industry
well deserves consideration. It is true that more difficulties are
involved in the manufacture of the soft than of the hard cheeses, but
the returns are correspondingly greater and quicker. Experience,
however, in other lines of dairy work shows that thorough scientific
study of the conditions, the problems, and the results sought is usually
rewarded by the development of successful practical methods of con-
trolling the entire process. The soft cheeses will l>e shown to be no
exception to this rule.
12 BUREAU OF ANIMAL INDUSTRY.
PRE8KNT MARKET CONDITIONS OF CAMEMBERT CHEESE.
There is good evidence that more than one type of cheese is placed
upon our markets under the name of Camembert. All of them have
sufficient similarities in texture and appearance to lead to classing
I hem together, but in flavor and odor they show marked differences.
Camembert cheeses secured from different regions show much varia-
tion. The study of the brands offered for sale in several cities empha-
sizes an entire lack of uniformity among them. The stage of ripen-
ing makes a marked difference in their appearance. One can find
upon the same counter cheeses in which ripening has barely become
distinguishable mixed with those in every stage of fitness and unfit-
ness for use, even to those which are absolutely putrid, and all are
likely to be sold as in prime condition for the table. Such cheeses on
the surface may be whitish or grayish in color, or yellowish to red-
dish, and may be dry or sticky with the accumulation of bacterial
growths. They may be practically odorless, or at least inoffensive to
smell, or they may emit very objectionable odors. These differences
are not due wholty to the degree of ripeness, for cheeses with very
different appearance are often found of the same texture and flavor
within. Moreover, cheeses with almost identical surface appearance
often differ decidedly in texture and flavor. The external differences,
we have learned, are due chiefly to the different kinds of microorgan-
isms that grow on the outside of the cheese, some of which are quite
unnecessary to the production, of the desired ripening. Consultation
with dealers shows that there is no accepted ideal among them for
Camembert cheese; they disagree constantly on such matters. There
also appears to be confusion in many quarters between the types
known as Camembert, Brie, and Isigny. This is probably due to the
fact that the imported Brie is ripened in very nearly the same way as
Camembert, differing only in size and shape and details of making,
while the type we have discussed as American Brie, which includes
the cheeses sold as Isigny, is very different in its appearance and
ripening process, as \vell as texture and flavor. It has been shown
that certain makers use these names as practically interchangeable
that is, regard them as the same general type of cheese molded at
different sizes and marketable at different degrees of ripeness, and
sometimes the same cheese exactly is sent to market bearing different
labels.
THE TRUE CAMEMBERT TYPE.
The true Camembert type of cheese is that imported from France.
These are shaped so that they fit wooden boxes about 4 inches in
diameter and li inches thick. Upon the removal of their wrappings,
the cheeses usually have a rather firm rind about one-eighth of an
inch thick composed of interlaced fungous threads supporting dried
cheese. Within, the ripe cheese should l>e a yellowish cream color, of
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 13
a waxy or creamy texture, often almost liquid in age, and with a dis-
tinctly characteristic flavor that is often not present in some American
types, even though they are labeled Camembert.
REASONS FOR SELECTING CAMEMBERT CHEESE.
Camembert cheese was selected as the first t} 7 pe of soft cheese to be
studied, for several reasons. It is one of the most delicate and expen-
sive of the soft cheeses, selling on the market usually for 40 cents a
pound. It has been succesfully produced by but one large factory in
our Eastern States. A few other factories have made cheese called
" Camembert," but their product has not made for itself a place in
our market in competition with the imported varieties. The market-
able period of this cheese is so short that the domestic brand should
have a great advantage over the imported. The shortness of the
ripening period four to five weeks together with the high price of
the cheese, insures a good return upon the investment and commends
its manufacture to the dairyman located too far from the city to sell
his milk as milk. Camembert cheese is already sufficiently well
known in this countpy to insure the ready sale of any brand which
shows itself to be always reasonably reliable. Considering the lack
of uniform quality in the Camembert cheeses seen on sale in half a
dozen of the large cities of the United States, the wonder is not at the
smallness of the demand, but that the people will continue to buy it
at all. Inquiry shows that many people are fond of the cheese, but
never buy it, lx?cause they can not depend upon the product. Dealers
do not like to handle it, for the samo reason. The fact that local de-
mand compels grocers in cities, often against their will, to handle such
cheeses is evidence that a really good and reliable Camenilxrt cheese
would make for itself a large market. It is readily seen, therefore,
that the making of this cheese offers good possibilities of being a
profitable industry. The demand could be more suitably met by
a large nurnlx>r of dairies or factories, each to supply a limited terri-
tory, rather than that an enormous quantity should be produced by a
few large concerns. That it requires special study to overcome the
difficulties of its production l>efore it can l>e made by the general
dairyman is also evident. For these reasons Camembert cheese
seemed to offer a most useful problem.
EUROPEAN LITERATURE CONSULTED.
When we started the work it was impossible to secure at once a
cheeso maker who was familiar with the factory manipulation in the
manufacture of these cheeses. We were, therefore, obliged to work
from the vague and inadequate descriptions of the details of making
them published in Kiiro]H'Hn journals. Practically no detailed
account of the practice of the lxst factories was to lx i found, while
the few scientific papers available scarcely touched the practical prob-
14 BUREAU OF ANIMAL INDUSTRY.
lems. Our problem was to imitate a market type without any per-
sonal knowledge of how it was produced. The most extended
scientific work that had been published was that of Epstein," who
made a somewhat careful study of the bacteria and molds found in a
series of ripe Camembert cheeses and endeavored to determine which
were the agents of ripening. The conclusion reached was that the
ripening of Camembert cheese is produced by the agency of bacteria
alone, although he also concluded that the ripening of Brie clm-r.
which is almost identical with Camembert, is aided by a mold.
In opposition to this there have been published the results of the
work of Roger, 6 who also studied the organisms concerned in the
manufacture of Camembert. He reached the conclusion that ripen-
ing is produced by the combined agency of a mold and at least two
different species of bacteria. Roger's work has the merit of having
been practically applied to the manufacture of Camembert cheese,
with results which he claims have been very satisfactory. But, on
the other hand, Roger has given the public no evidence of the work
he has done to reach his conclusions. He has preferred to develop
the practical side of his studies as a trade secret, and has, therefore, in
his published articles withheld all details of his experiments and
descriptions of his molds and bacteria. Consequently the published
account of his work is less useful than that of Epstein. Neither
worker has given sufficient data to demonstrate his claims or prac-
tical methods of cheese making.
The fact that the conclusions of Epstein so totally disagreed with
those of Roger suggested that the subject at all events demanded
further and more thorough investigation. Moreover, it will be seen
in the following pages that our own results, in a large measure, differ
from those of both of these investigators, for we have been able to
produce Camembert cheese, typical both in flavor and in texture, by
a method involving organisms essentially different from those claimed
by Epstein or Roger as the cause of ripening. It is scarcely possible
that the details of either set of experiments can be all mistakes. The
fact that three series of studies upon the same subject have reached
three quite different results indicates one of two things: Either the
cheeses used in the three series mentioned were of really dissimilar
types, though under the same name, or a very similar final product
can be produced by different agents. Possibly both of these explana-
tions are needed to account for the discrepancies. Certain of our
experiments have yielded information which" leads to the belief that
each of the .three lines of experiment has brought out part of the
truth. These results will be discussed in detail later.
Archiv f. Hyg., Bd. 43, pp. 1-20 ; Bd. 45, pp. 354-376.
& Sep. pub. Soc. Fran. d'Encore, Find. Laitiere, 1902, and Uevue Hebdoma-
daire, 11, pp. 3U7-343.
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 15
OUTLINE OF OUR METHOD OF PROCEDURE.
The first problems which we undertook to study were scientific
rather than purely practical. The reasons for this were two. First,
we were unable at the beginning of our work to secure a cheesemaker
familiar with the factory methods of making and curing this type of
cheese. Second, it was not our purpose merely to imitate the methods
in use in producing this cheese in France. This would have been
hardly worth while unless we could obtain a thorough understanding
of all the principles involved. It was desirable, therefore, at the
outset to conduct a series of experiments which would l)ear directly
upon the determination of these principles, and this could be best
clone by making a few cheeses under controlled conditions. In this
way we reached conclusions upon the problems involved which stood
the test of practical experiment and which led quickly to successful
results when cheese making was actually begun.
After securing the services of Mr. Issajeff we began the application
of our scientific studies to the practical question of cheese ripening.
No description of the practical method of making and caring for
these cheeses will be given in this introductory paper. These will be
given in a later bulletin to follow shortly. It will be understood,
however, that the manufacture of Camemlxrt cheeses is being now
actually carried on at the Storrs Experiment Station, and that the
scientific results given here have been actually applied to the process
of manufacture.
RIPENING OF CAMEMBERT CHEESE.
A brief outline of the ripening of Camembert cheese is as follows:
The rennet curd is first ladled into forms, filling them up to a depth
of alxnit 5 inches, and is allowed to stand for several hours for drain-
ing. During this time the curd settles until it reaches a thickness of
alxMit 2 inches, when the cheeses are turned and allowed to stand in
the forms. During the next twenty-four hours the curd settles still
further, reaching a thickness of H inches. The first change a fleet ing
the rij>ening is the souring of the curd, which logins while the cheese
is in the forms. In the course of two days the curd has Ixvome hard
and sour. The cheeses are then removed to the ripening cellar.
After a few days in the cellar the molds lx\gin to grow upon their
surface. From this time until the end of the ripening period there
is little visible external change in the cheese, except in the luxuriant
growth of molds and in the fact that during the later stages of the
ripening there commonly appears upon the surface of the cheese a
brownish-red growth, which is likely to be moist and slimy. This
brownish-red growth has IHMMI regarded as a necessary factor in the
ripening of cheese, but whether it has any necessary connection with
it we are as yet uncertain. During the ripening the cheeses at first
16 ItfKKAIT OF ANIMAL INDUSTRY.
ixcome somewhat hard and resistant when pressed with the finger.
Later they soften noticeably, and when they approach ripeness they
are quite soft and yield readily to the pressure of the finger. The
ri|Kmss is determined chiefly by the softness to the touch and is
easily told by experience.
The ripening changes are studied by cutting sections of the cheese
at various stages. The photographs (see plate 2) show such stages
of ripening, of which a brief epitome is as follows:
After :il"'in two weeks the acidity of the curd begins to be notice-
ably less, especially at the surface, and as the ripening progresses
the aciditj' in time disappears so far as indicated by the test with
litmus pa|Hr. Meantime a change in the texture of the curd is evi-
dent to the eye. Beginning at the outside, there appears a change of
the hard curd into a somewhat softer, waxy material. This layer of
waxy cheese. beginning at the surface, slowly extends toward the cen-
ter and in the course of four to five weeks should reach the center. In
half-ripened cheeses there will be found a layer of softened con-
sistency on the outside, with a central mass of acid curd within, which
has not yet been affected by the ripening agents. When the cheese is
completely ripened the softening extends to the center, and the
whole cheese, from the outside to the middle, has assumed a con-
sistency of moderately soft butter. When in prime condition the
cheese is soft enough to spread upon bread or crackers, but not soft
enough to run. An over-ripened cheese, however, becomes still softer
until in time the whole interior of the cheese below the rind is con-
verted into a nearly liquid consistency, which will run out of the
chcvsc readily if the rind is broken. On the other hand, an under-
ripened cheese will show more or less of the sour curd in the center
which has not been affected by the softening agents. The cheeses pur-
chased in the market are very frequently in one of these two condi-
tions, either overripe or underripe. (See plates 1 and 2.)
During the ripening process there appears a peculiar flavor which
is characteristic of this type of cheese. This flavor is sometimes not
noticeable until the cheese is nearly ripe, so that a cheese which is two-
thirds ripened may lack the flavor in question. The final result is a
cheese with a firm, moldy rind, and with the contents uniformly soft
to the center, and possessing a characteristic, piquant flavor which is
found in this type of cheese only.
The problems to be determined were the causes of the phenomena
of ripening, and these are manifestly three.
(1) The cause of the original souring of the curd.
(2) The cause of the decrease in the acid and the gradual softening
of the curd.
(3) The development of the flavors.
But. No. 71, B. A. I.
PLATE 1,
FIG. 1. A BOARD OF CAMEMBERT CHEESES.
FIG. 2. AN IMPROPERLY RIPENED CAMEMBERT CHEESE.
The outer portion is liquelied, while the center is hard, sour curd.
A SERIES OF CAMEMBERT CHEESES.
Fig. 1, unripened cheese; fig. 2. half-ripened cheese with well-developed moldy rind;
lig. 3, cheese two-thirds ripened but with centrul mass of hard curd showing some gas
holes; tig. -1, cheese nearly ripe and so soft that the upper edge moved because of its
weight during the exposure.
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 17
COMPENDIUM OF OUR SCIENTIFIC INVESTIGATIONS.
Our work has been developed along three scientific lines; the results
of each line are summarized below. More detailed accounts of these
studies will follow in separate papers.
MYCOLOGY.
The constant appearance of mold in or upon soft cheeses such as
Roquefort, Camembert. and Brie has been referred to already. The
works of Epstein and Roger, already discussed, give two views of their
relation to the production of cheese. Epstein believes the mold
entirely unnecessary in the production of Camembert, but notes its
presence and important function in the ripening of Brie. Roger,
on the contrary, believes the mold to be really essential in the ripen-
ing of Camembert. Other investigations published record the con-
stant appearance of mold upon Brie, but give little or no information
with reference to Camembert. Popular descriptions of Camembert
cheese factories, however, describe the constant presence of a special
mold in the ripening cellars. The references to molds in the litera-
ture of the subject and their constant presence upon the cheeses fixed
for us a series of questions: What molds are associated with Cam-
embert cheese as found in the market? Which, if any, of these are
essential to its proper ripening? Exactly what changes in the
ripening process result from fungous activity? What molds are
deleterious? And what is their action?
With these questions in mind we first made a cultural study of
the flora of the Camembert cheeses found in the American market.
In this about twenty molds were soon found, isolated in pure cul-
tures, and descril>ed. A more careful study of our cultures with a
comparative examination of the surfaces of the different kinds of
cheese showed that only a few of these molds were really common,
while but two of them were always present.
Having the organisms in pure cultures, we next undertook to
determine what relation, if any, each might lxar to the ripening
process. Since it is practically impossible to make cheeses entirely
free from other molds, or bacteria, or both, without changing their
chemical nature, some controlled means had to lx devised to studv
the various agents of ripening and the steps of the process. We
found it necessary to limit this comparative study to the effects of a
small group of molds. These were grown in pure culture under
known conditions upon milk, casein, curd, cheeses, and sjM'cial media,
all so devised as to determine separately different phases of the
physiological action of the molds. For this study we used (1) the
one we shall call the 4 * Cameml>ert mold" (culture No. 1*28); (-2)
18 BUREAU OF ANIMAL INDUSTRY.
the related species isolated from Roquefort cheese, which we will call
the" Roquefort mold;'' (3) one species of Mucor; (4) O'idium lacfis,
and sometimes other species of the same genus, Penicillium. In some
studies the list has been extended to a dozen species in each experi-
ment, because with the larger numbers comparison of characters
gives a letter judgment of the real nature of the processes studied.
The souring of the curd as a consequence of the presence of lactic
bacteria has already been descrilxnl as the first change after the cheese
is made; also the fact that in Camembert- cheese the lactic acid so
produced later largely disappears, so that when fully ripe such a
cheese commonly has an alkaline reaction to litmus. This disap-
pearance of acid, together with the well-known ability of many molds
to reduce acidity, has been regarded as the main, if not the only,
function of the mold in ripening cheese. This view has been widely
quoted. Comparative cultures of a large number of species show that
a very large proportion of the common molds are able to neutralize,
if not decompose, lactic acid, but they do this in different degrees and
at very different rates. The Camembert mold increases the acidity of
the medium during its first few days of growth, then changes and
reduces acidity for the remainder of the period of activity. Some
other forms reduce acidity from the first and do it much more rapidly
than this one. If this were the only function of the mold in cheese
ripening, many other species should be equally valuable, or by work-
ing more rapidly would probably be far more effective. This part
of the ripening then might be accomplished by any one of a large
number of forms.
We then turned our attention to determining whether any or all
of these fungi acting in pure cultures possessed the power to change
the hard curd to the semiliq"uid or viscous condition of the ripened
Camembert cheese. A long series of cultures upon sterilized milk,
milk agar, and sterilized curd were prepared. Careful bacterial
determinations were made to show that the cultures were pure. From
these the following facts appeared : Sterile milk is slowly but almost
completely digested by nearly every species of fungus tried. Flakes
of curd in milk agar plates disappear as a result of the action of the
fungi selected for this experiment. Masses of curd put in flasks and
sterilized in the autoclave until they seemed almost the texture of rub-
ber became nearly completely soluble in water after four to six weeks'
action of either the Camembert or the Roquefort mold. Experimental
cheeses made from time to time, with the addition of large quantities of
"lactic starter" to exclude other bacteria, were inoculated with this
mold and carefully kept pure in bell jars. Under favorable condi-
tions such cheeses assumed the texture of ripe Camembert cheese in
from three to five weeks. Repeated analyses have shown that this is
approximately the same kind of digestion as is found in the ordinary
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 19
Camembert cheese. These pure-culture experiments have thus shown
that this species of mold possesses the ability to change the texture of
curd to the texture of ripe cheese in the period of time recorded by
the makers of such cheese as necessary for the ripening of Camembert.
How, then, is such a result accomplished ? Johan-Olsen in describ-
ing the Swedish " Gammelest " declares that " he who eats Gam me -
lest eats truly more mold than cheese." In other words, he believes
that in its ripening process the changes are due to the complete pene-
tration of the cheese by fungous threads which act directly to change
its character. Gammelest would then be fungus flavored with cheese !
But cultural studies and careful microscopic examination of hundreds
of sections from a number of different Camembert cheeses show that
our problem is very different. In Camembert cheese the mold forms
a felted mass of hyplue upon the surface and penetrates, perhaps,
one-sixteenth of an inch into the curd. It never reaches even one-
third of the distance to the center of the cheese, while its spores are
borne only upon the surface. This, together with a very thin outer
layer of curd, constitutes the rind of the cheese.
Further, our experiments referred to above show also that on sterile
milk the mold forms colonies floating upon the surface only, while a
mass of curd in the bottom of the tube, fully 2 inches away from the
nearest fungous thread, may still continue to be digested. It is clearly
impossible to attribute these marked digestive effects at such distances
to the direut action of the fungous threads upon the medium. From
similarity to other known processes, the presumption arises that these
are the result of the secretion of enzymes by the mold.
To test the validity of this hypothesis, a set of cultures were grown
for a long time on Haulm's fluid, then examined by the chemist. The
presence of an enzyme capable of digesting proteid was satisfactorily
demonstrated. We have, therefore, shown by pure-culture methods
that the Camembert mold (PeniciUium candidumf) is not only capa-
ble of changing the acidity of the curd, but is able also to cause such
changes of the curd as will account for the texture of the ripe cheese,
and that this result is due to the secretion of an enzyme.
A cheese ripened by this mold alone is white, soft, creamy, and
entirely palatable, but is wanting in color and completely lacks the
jHvuliar flavor for which Cameml>ert cheese is sought in the market.
After repeated tests had shown the same result, we U'gan to seek for
some other organism capable of producing the desired flavor. The
discovery of this flavor in certain experimental cheeses at this time
was followed by their immediate microscopic examination, which
showed the presence of the second of the two fungi originally found
on CamemlM-rt cheese. This mold, the well known and universally
distributed Onl'mm (or (toxjutrn} luctix had l>een discarded from
much of our work on account of some failures and objectionable
20 BUREAU OF ANIMAL INDUSTRY.
results. We now found that the inoculation of this organism upon
cheeses partially ripe and lacking flavor would lead to the pro-
duction of the flavor distinctly in a very few days. Although from
its habits of growth its development upon cheese is nearly always
accompanied by a rapid multiplication of bacteria, this seemed
good circumstantial evidence that Oidium lactis has something to do
with the flavor of Camembert cheese. Additional evidence is found
in the fact that the examination of proj>erly flavored ripe cheese of
our own make and from the market has never failed to show its
presence, and it has never been shown to be present upon ripe cheeses
without the flavor being present also. The great difficulty surround-
ing complete proof of the agency of a particular organism in pro-
ducing flavor lies in the fact that the flavors do not begin to appear
until the acidity of the curd is much reduced and the digestion
attributed in our work to the Penicillium has advanced considerably.
Flavors, the chemists tell us, are associated with the " end products
of digestion." When ripening has reached such a stage as to permit
flavor formation, the growth of O'idium upon the surface of a cheese
usually becomes complicated by the development of surface bacteria,
which becomes possible at about that time on account of the reduc-
tion of the acidity. A final conclusion as to whether or not 0'idium,
lactis alone produces the flavor will depend upon an exhaustive test
of those bacteria so constantly associated with it.
The bacteriologists, as will be seen later, have shown that few
bacteria, except lactic species, are found within the cheese, and none
of them seem able to produce flavor. The production of flavor by
bacterial action would then depend entirely upon such bacteria as
might grow upon the surface of the cheese. In addition to the evi-
dences already mentioned, it is found that a cheese entirely covered
by a good growth of the Camembert mold evaporates water rapidly
and develops a hard, dry rind, so dry as to obstruct the entrance of bac-
teria from without. In certain of such cheeses, into which 0'idium
was inoculated at the time of making, the characteristic flavors
finally appeared, Avithout the association of the reddish slimy surface
so commonly seen. Further, many cheeses made and ripened entirely
without 0'idium lactis have in their later stages become covered with
bacteria, which produced the surface appearances so often described,
but failed entirely to develop the typical flavor. In the light of the
bacteriological proof that the interior of the cheese is practically a
pure culture of lactic organisms, the production in some cases of both
the flavor and texture of properly ripened Camembert cheese without
a rich surface growth of bacteria is very good circumstantial evi-
dence that Oidium has some function in producing the flavor.
A cheese inoculated with Camembert Penicillium will not begin to
show delicate w r hite threads of mold for about three da vs. The mold
CAMEMBEBT TYPE OF SOFT CHEESE IN UNITED STATES. 21
develops rapidly from that time until, by the tenth day (sometimes by
the seventh or eighth), the cheese is usually covered entirely by a pure
white, cottony mass of threads forming a layer possibly one-eighth
of an inch deep. With the ripening of the spores or conidia the color
gradually changes to a greenish gray. This change is completed
during the third Aveek usually, and no growth seems to take place
afterwards. During the later stages of mold growth large drops of
water are excreted by the mold and evaporate from the surface. The
mold, therefore, has a rather definite period and course of develop-
ment. During that time it seems to be so thoroughly adapted to
thrive upon cheese as to exclude almost every other form, but after its
cycle of development is complete it may be followed by other species
which, if present before, are held in check by the more vigorous
species.
One more phase of the mold problem perhaps belongs in this pre-
liminary paper. Efforts have l>een made to find whether it is pos-
sible to substitute other varieties or species of the same genus of
fungi for the one we have been using. For such studies the mold
of Roquefort cheese has been carried as a check upon our work in
numerous experiments. One other variety oi Penicillium, differing
only in that it never changes color, but remains pure white, has been
tested. Cheeses have been made with these three and several other
related species. The Roquefort Penicillium grows strongly and
rapidly upon cheese, but its presence always causes a bitter taste,
which remains pronounced during the period of four or five weeks
necessary for the ripening of Camembert. So strong is this effect
that the presence of a colon}' of this mold less than an inch in diam-
eter may often Ix? detected by the taste of the cheese 2 inches away.
The pure white Penicillium in its ordinary reactions to culture media
seems to be identical with the Cameml>ert mold, but when tested
upon over one hundred cheeses produced a texture so entirely different
from it as to make the product entirely worthless. It seems most sur-
prising that two forms so closely related in structure and in every
reaction studied should produce such different results in cheese ripen-
ing. Several other forms have been tried. Some produce pigments
which discolor the cheese; some produce bad flavors. A set of four
species, although allowed to grow for a period of six weeks, produced
no softening of the curd. In two of these cheeses especially it was
found that the curd was still sour and little changed less than one-
fourth of an inch U'low the colonies of the I'lingus. The four cheeses
used in this experiment belonged to a set of twenty, sixteen of which
rijxMied readily in tin- ordinary way. Sets of cheeses made from the
same milk and treated alike ripened quite uniformly, as a rule. Of
this set, 1(5 ripened in the usual manner when acted upon by Cameni-
lx>rt Penicillium. Four inoculated with other Penicillia failed to
22 BUREAU OF ANIMAL INDUSTRY
soften. Such a result shows that neither the molds used nor the
bacteria and enzymes otherwise present were capable of producing
the texture of Camembert in four cheeses of a series, the remainder
of which were readily ripened normally.
- BACTERIOLOGY.
The universal presence of bacteria in all dairy products suggests
of course that they may play an important part in the ripening of
cheese. In the various scientific accounts of Camembert cheese
bacteria are always mentioned, and both Epstein and Roger attributed
to their action a considerable part of the ripening and flavor produc-
tion. Our own preliminary studies of the market cheeses always
showed the presence of bacteria in great number and in considerable
variety. It was manifest that most of these were purely incidental
organisms not contributing at all to the ripening. In the study of
these bacteria the following questions have been kept in mind: Which
kinds of bacteria are always found and which are only occasionally
present? Which kinds of bacteria are present in fresh cheese?
Which kinds of bacteria grow during the ripening? Do any of these
bacteria produce the Camembert flavor? Do the bacteria contribute
to the chemical changes that occur during the ripening? Part of
these questions we have already answered, but part of them require
further study.
We first made a bacteriological study of a considerable number of
ripened cheeses, isolating all species of bacteria that could be found.
Except that lactic bacteria were always found, this work gave no
significant results. We were somewhat surprised to find that the
group of bacteria that digest proteids the liquefiers were commonly
entirely absent. Inasmuch as the cheese is so evidently a digested
product, we had anticipated finding them in large numbers. This
absence in large numbers suggested that the digestive agent must be
looked for elsewhere than in bacterial action.
Our search for a flavor-producing bacterium has hitherto been
equally unsuccessful. Epstein states that by inoculating into a casein
solution one of the bacteria he isolated from Camembert cheese he
obtained a typical flavor. We have been unable to find such a
bacterium. We have carefully studied the action upon milk of the
organisms isolated from cheeses in the hope of finding a flavor-
producing bacterium, but have thus far been unable to find one. We
have, however, found bacteria that produce flavors that remind us
of the American Brie, and of other flavors associated with some sam-
ples of cheese, but none that gave the typical Camembert flavor. We
are at present inclined to think that this flavor must be sought in
some other line, as has been shown in the previous pages.
The study of the growth of bacteria in the cheese during the
OAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 23
ripening is very interesting and significant. We have studied this
problem in cheeses made with and without a lactic starter. The
cheeses made with a lactic starter are more uniform in ripening.
Experience has shown that we can with greater certainty obtain a
first-class cheese if we use such a starter. For this reason the stud} 7
of the bacteria in such cheeses is especially significant. The results
are very simple, for they are an exact repetition of what ocelli's in
souring milk, as has been already shown. At the outset there is
commonly a small variety of bacteria present, which develop for a
short time, but in a few hours the lactic bacteria get the upper hand
and all other species practically disappear. After about two days
the cheeses contain a practically pure culture of lactic bacteria.
These increase in numbers until they are about 000,000,000 per cubic
centimeter, and then remain at about this number during the ripen-
ing, somewhat falling oif toward the end. In these cheeses there is
no growth of liquefying bacteria to which might be attributed
such digesting action. There are commonly a few of these at the
start, but they never multiply so that they become abundant enough
to lead to the supposition that they are likely to play any part in the
ripening. Moreover, as mentioned above, the ripening of these
cheeses proceeds from the surface inward, a fact that indicates that
the ripening agents are chiefly or wholly on the surface.
These facts are rendered more significant by the bacteriological
study of one set of 20 cheeses that were made without lactic
.starter. In these cheeses the lactic bacteria never obtained such a
complete ascendency over the other species as where a starter was
used. While they l>ecame very abundant we found that certain lique-
fying bacteria multiplied and remained inconsiderable numlxrtothe
end of the ripening. The presence of large numbers of such enzyme-
produeing bacteria during the ripening suggests that they may have
contributed to the process. It was especially significant to find (1)
that this cheese ripened in the center l>efore it did on the outer edge,
as would !H> expected if the bacterial enzymes aided in the ripening,
and ( % J) that the final product was in flavor and texture not very dif-
ferent from the more typically ripened cheeses. This would seem to
suggest that while the normal rijMMiing is not due to the bacteria or
enzymes in the body of the cheese, but to the organisms on the sur-
face, under some circumstances the former may materially contribute
to the pnx-ess, ami while producing a somewhat abnormal rijXMiing
will give in the end a similar result.
From the facts thus outlined \ve have reached the conclusion that
the bacteria in the l>ody of the cheese are in normally ripening cheeses
concerned only in the souring of the curd as preparatory to the later
changes. This leads naturally to the study of the organisms present
" Cuiin :inil KsttMi. Kept. Storrs St:i. l!KM-2.
24 BUREAU OF ANIMAL INDUSTRY.
in the outer layers. This rind contains a variety of molds and
bacteria. The presence of certain molds in the rind of every Camem-
lxrt cheese examined suggested from the first that they must be
important agents in the ripening process. This has been demon-
strated in the preceding section, but this rind also under most cir-
cumstances contains many bacteria. Our study of these bacteria
and their actions is not yet complete. We have as yet no good
evidence that their action is necessary to the production of a first-
class product. That they occasionally modify the final cheese is
quite certain, and we are also convinced that they play an impor-
tant part in the ripening of the type we have called the "American
Brie." Whether they have any influence on the production of the
typical Camembert flavor can only be determined by subsequent
experiments.*
It should finally be stated that the surface of the ripening cheeses
very frequently becomes covered by an abundant growth of yeast,
a phenomenon which the maker's call " sweating." At present we
have no grounds for believing that this has any influence upon the
ripening unless it be to check in a slight degree the drying of the
surface of the cheese.
CHEMISTRY.
The chemical composition of a normal Camembert cheese is quite
well known, considerable work in this line having been done. This
work, however, is not of much help in studying the changes that take
place during the ripening process and makes it necessary to go over
the same ground that has been covered in the work done on Cheddar
cheese.
The great difference in the making and ripening of Cheddar and
Camembert cheese soon made it apparent that an entirely different
set of factors were at work and that the chemical side of the inves-
tigation opened a field which would require considerable time to
complete.
We have undertaken (1) to follow the chemical changes which
render the insoluble curd of a newly-made cheese into soluble forms
Since the above was put In type we have received from France a set of
cheeses selected and forwarded by M. Georges Roger, accompanied by a letter
giving data as to age and condition of ripeness. Studies from the surface of
these cheeses have shown the presence of the same Penicillium we are using,
of Onlium lactix, and of the same species of bacteria which apj>ear on our
cheeses and contamination with the same other molds we have met. These
bacteria on the older cheeses formed reddish-brown slime, almost entirely cov-
ering the cheese. Systematic tests of these red-slime producing bacteria are
being made to ascertain what effect they may really have UJKHI the ripening of
the cheese. Cultures made from the interior of the cheese showed practically
pure cultures of lactic organisms, ns in our own cheese. These results suggest
that we have the same organisms as M. Roger, though under different names.
OAMEMBEKT TYPE OF SOFT CHEESE IN UNITED STATES. 25
found in a cheese ready for consumption ; ('2) to isolate and study
the enzyme or enzymes which cause these changes; (3) to investi-
gate the changes which concern the production of the peculiar flavor
which is characteristic of Camembert cheese; and (4) to show the
relation which each of the constituents of a newly-made cheese bears
to the ripening process.
Up to the present time all our attention has been given to work
upon the nitrogen constituents of the cheese. This was because of
an early belief, confirmed later to some extent, that the fat plays a
secondary part in the ripening and in the development of the flavor
in Camembert cheese.
The market product was first studied to establish a standard. This
was secured from the analyses of a number of imported cheeses.
Our cheeses compare favorably with these, yet we hope to obtain im-
provement by a slight modification in our methods of making and in
the handling of the cheeses in the ripening cellar that is, acidity,
humidity, temperature, and washing the cheeses.
TABLE 1. Chemical changes in the ripening of Camembert cheese.
Labora-
tory No.
Age.
Total ni-
trogten.
Nitrogen
as mono-
lactate.
Nitrogen
as water-
soluble.
Nitrogen
as amids.
Nitrogen
as am-
monia.
D<iys.
Per cent.
Per cent.
Per cent.
Per cent.
Per cent.
25a
1
2.73
0.40
0.32
0.23
0.00
25b
3
2.73
.41
.41
.21
.00
26a
5
2.81
.39
.45
.21
.00
25c
6
2.73
.41
.54
.27
.Of)
2Bb
7
2.81
.:
.52
.27
.00
27a
8
2.92
.43
.68
.3t(
.00
28c
10
2.81
.87
.73
.38
.00
27b
11
2.92
.2
.75
.111
.00
28a
12
2.90
.SI
.
.53
.00
27c
14
2.92
.57
1.66
.48
.00
2Kb
18
2.90
.27
1.78
.4S
.01
29a
19
2.65
.34
1.59
.43
.02
28c
20
2.90
.17
1.85
.42
.(Mi
a29b
21
2.66
.13
1.79
.48
.OB
14
28
2.61
()
2.0
.40
.21
a This aeries of cheeses " went bad " with bacteria af Ur this sa i u i .1. was taken. For coiupari-
non the analysis of another rheeae, No. 14, is given.
t> Not determined.
The first change is the precipitation of the casein of the milk by
rennet. This precipitate or coaguluin carries down with it most of
the fat and some of the milk sugar. The sugar is changed in a few
days by bacteria into lactic acid, rendering the curd distinctly acid
in reaction to litmus. The fat undergoes some slight changes, but
the precipitated casein undergoes a very marked modification.
Practically no change takes place in the cheese except the develop-
ment of acid during the first twelve days. Then the curd Ix-gins to
20 BUREAU OF ANIMAL INDUSTRY.
soften and is gradually changed from an insoluble lump into a soft,
creamy mass almost entirely soluble in water. This change start-
at the outside and progresses inward toward the center until tin-
whole mass has been acted upon.
This change in the curd is probably due to one or more enzymes
produced by the molds growing on the cheese. These enzymes begin
their activity at about the fruiting time of the molds about the
twelfth day. They have not been separated as yet, but the evidence
at hand seems to indicate the presence of both tryptic and peptic-like
enzymes.
Some idea of the changes that take place can be obtained from the
table on the preceding page.
GENERAL CONCLUSIONS.
Although the results of these lines of work are still incomplete,
some conclusions have been reached which bear directly upon the
practical problems. These conclusions have been put to the test of
constant use since the establishment of our ripening cellar, and have
brought a much better understanding of the problems involved than
was before available.
It has been conclusively shown that the making of Camembert
cheese is not dependent upon unique conditions obtainable only in
very restricted localities, but rather upon securing the proper cul'tures
and conditions, which are possible almost anywhere. Cheeses made
at this station have been compared with the best imported goods
obtainable in America, and pronounced practically identical, not only
by members of the force, but by importers and connoisseurs who have
eaten this type abroad. These cheeses are in no way inferior in tex-
ture or flavor to the best market article. They have been ripened by
inoculation with pure cultures and have been watched constantly and
tested by the bacteriologists, the chemist, and the mycologist at every
stage of their ripening. The element of chance is thus excluded from
the result. The organic agents involved are known. The difficulties
encountered in the attempts to establish this brand of cheese-making
in the past have, therefore, been due to insufficient knowledge of the
molds and bacteria involved in the process, rather than to failure in
obtaining any peculiar conditions or the proper milk.
Our practical work also proves that it will be possible to control
the ripening to such an extent as to produce a more uniform product.
It has been the practice of the makers in the past to provide suit-
able conditions and then, as one has put it, let u nature do the ripen-
ing." As a consequence of this " hit or miss " way of doing things,
some cheeses are inoculated with the proper cultures from the first,
some bear one organism and not the other necessary, while wrong
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 27
s juries of bacteria and mold very commonly grow in or upon the
cheeses, so that the final results are very uncertain. This accounts
in a large measure for the variation in appearance, texture, and
flavor of such cheeses, and is apparently responsible for the differ-
ences between the imported and American-made types.
CONTROL OF THE BACTERIA AND MOLDS.
It became evident very early that uniformity depends upon the con-
trolling of the growth of bacteria and molds from the time the cheese
is made until the close of the ripening. This pointed to inoculation as
the best means of reaching the desired result. Experience has justi-
fied this conclusion at every step. The use of lactic starters in butter-
making had already established a simple means of controlling the
development of bacteria. The control of the molds was a more
difficult problem, inasmuch as there are at least two different molds,
and to obtain the desired result depended upon keeping the proper
balance between these two organisms growing upon the same surface,
but having different functions. AA T e have learned that by varying
the conditions and growth of the Penicillium we can, to a large
extent, control the development of the other mold. Oidium will
not grow profusely upon a cheese already covered with the white
mold, and by developing a luxuriant growth of the latter we can
hold the former largely in check. Cheeses in which the Oidium
has not developed abundantly are either mild in flavor or lacking
entirely in the typical Camembert flavor. We have succeeded in
producing cheeses perfectly ripened so far as texture is concerned,
but flavorless, others with the flavor distinctly developed but mild,
and others again with a strongly developed Camembert flavor. Our
experiments thus far seem to show that by controlling the growth
of these two molds we can control both ripening and flavor produc-
tion, although, as almve suggested, we have not yet absolutely ex-
cluded the possibility that bacteria may play some part in the process
of flavor production.
The important practical problem, then, appears to be to devise
methods of treatment that will enable the cheesemaker to control the
growth of microorganisms, and primarily the lactic bacteria and
two species of molds. The control of the first process of souring of
the curd is very easy. It is only necessary to apply here the method
now so widely used in cream ripening, namely, the inoculation of tin*
milk with lactic starters. These starters, if placed in fresh milk,
insure a proper souring with perfect uniformity.
The control of mold growth is a more difficult matter. l>ccause of
the great likelihood that the cheeses in handling will become inocu-
lated on the surface with other than the desired organisms. The
28 BUREAU OF ANIMAL INDUSTRY.
solution of the problem, however, appears to be found in proper
attention to throe factors.
(1) The inoculation of the cheese at the proper stage with a com-
paratively large quantity of the spores of the desired organisms.
At just what stage the inoculation should be made and in what way
we are not at present prepared to say. The question of direct inocu-
lation of cheese with two molds as a means of controlling ripening
is an entirely new one in the discussion of this cheese problem. The
practice of the factories has been to allow the cheese to develop the
1) roper mold after entering the ripening room. But there seems to
be .a good reason for a change in this particular. Milk as it reaches
the factory already contains the germinating spores of many species
of fungi. Some of these forms develop even more rapidly than the
ones essential to our purpose. The proper mold has never been found
by us in thousands of cultures of milk as it comes from the barn,
while undesirable species are very common. If the Penicillium
essential to cheese ripening is to take and maintain the ascendancy
in the ripening process it is imperative that it should begin to grow
as soon as the cheese is made. Although further experience is neces-
sary to determine the best time and manner of the introduction of the
molds, we are convinced that their early introduction gives advan-
tages more than compensating for the additional labor involved.
(2) The cleanliness of the ripening cellar. The cheeses remain in
the ripening cellar from four to five weeks, where the conditions are
kept ideal for mold growth. If the walls, shelves, or floors are
allowed to become covered with a growth of miscellaneous molds,
their spores are sure to reach the cheeses and are then liable to pro-
duce trouble. The remedy for this is cleanliness, and special care
should be taken to prevent the growth of molds in the ripening room.
(3) The treatment of the cheese in the cellar. We have learned
that variations in moisture and temperature of the ripening cellar,
and different methods of handling the cheeses, all greatly influence
the growth of the molds and bacteria. A perfectly uniform method
of handling is therefore necessary for perfect uniformity in results.
Differences in little matters of detail are found in the treatment of
cheeses in different factories, and these become the basis of each
maker's brand. Some wash their cheeses ; others do not. Some turn
them more often than others. Some send them to the market in a
greener condition than others.
GENERAL SUMMARY.
The conclusions we have drawn at present are as follows:
1. The ripening of Camembert cheese is due to the presence of
definite molds and bacteria. It is not merely a matter of manipula-
tion and locality.
CAMEMBERT TYPE OF SOFT CHEESE IN UNITED STATES. 29
2. One mold (Petiicillium candid um?) produces the principal
changes in the curd which gives the texture of the normally ripened
cheese. Another (Oidiurn lactis), acting in conjunction with the
first, is necessary to produce the flavor.
3. The presence of lactic bacteria (commonly used in the form of a
starter) produces the necessary acidity of the curd and prevents
further bacterial action within the body of the cheese.
4. No other organisms seem absolutely necessary to produce the
texture and flavor of Camembert cheese, though other species of
bacteria are always present in or on the cheese.
5. It is possible so to control the process of ripening as to produce
such results with reasonable uniformity.
6. It is perfectly practicable to produce first-class Camembert
cheese in the United States.
ANOTHER REPORT PROPOSED.
In this introductory paper only the general principles can be out-
lined, and we hope to follow this paper soon with another which
will give in detail the actual methods of making and handling which
we have found most successful in producing the best product.
o
001
I II I II II 1 1 II
102566 5
o
3 -* x =
c
CD
O
=l[lf
-1 '{:
m 2
LTVnJ
i 1 !"?
Ifti!
O
co
LO
c_n
11
1 =5
I 7>
8>
LrunJ
*^
K^
