KEPRODUCTIV.G CYCLE IN DASYURUS VIVBRBINUS. 1 3 3

The Reproductive Cycle in the MarsupialDasyurus viverrinus.

By

J. P. Hill, D.Sc.(£din.),

Jodrell Professor of Zoology and Comparative Anatomy,

and

Clias. H. O'Donoglme, D.Sc.(Lond.),Beit Memorial Fellow.

(From the Zoological Laboratory, University of London,University College.)

With Plates 6 to 8.

CONTENTS.PAGE

I N T R O D U C T I O N . . . . . 134

M A T E R I A L . . . . • • 135

ANCESTRUS . . . . . . 136

PRO-OESTRUS . . . . . - 1 3 9

Uter ine Changes . . . . . 140( E S T R U S . . . . . . 142

Uter ine Changes . . . . . 143POST-CESTRUS . . . . . 144

Uter ine Changes . . . . . 145Ovulation . . . . . . 147

P R E G N A N C Y . . . . . . 14SGestat ion Per iod . . . . . 149General Changes . . . . • 150Ear ly P r e g n a n t Uter i . . . . .1151

N U R S I N G P E R I O D . . . . . 152

1 3 4 J . P. HILL AND CHAS. H. O ' D O N O G H U E .

P A G E

P S E U D O - P R E G N A N C Y . . . . . 153

Uter ine Changes . . . . . 154METCESTBTJS . . . . . . 1 6 0

S U M M A R Y . . . . . . 160

C O N C L U S I O N S . . . . . . 165

1. Comparison of (Estra l Cycles in the Metather ia andEu the r i a . . . . . . 165

2. The Monoestrons and Polycestrous Conditions . . 169L I S T O F R E F E B E N C E S . . . . . 1 7 1

D E S C R I P T I O N O F P L A T E S . . . . . 174

I N T R O D U C T I O N .

IN receut years a great deal of attention has been paid tothe reproductive processes in the Mammalia, and in particularto the phenomena connected with the cestral cycle. Observa-tions relating to the latter, however, have been almostentirely based on Eutherian mammals, and, indeed, largelyon Primates (Heape (8, 9, 10), Van Herwerdeu (12), Hitsch-mann and Adler (15), and others) and various domesticatedor semi-domesticated Entheiia (e.g. sheep (16), dog (19),ferret (17), Marshall). Having at our disposal a large volumeof records relating to the breeding habits of Dasyurus, aswell as an abundant supply of material, we have thoughtthat an inquiry into the reproductive cycle in this member ofthe Marsupialia (in many respects a more primitive groupthan the Eutheria), might not only be of interest, but might,perchance, throw light on some of the problems relating tothe Eutherian cestrous cycle which still await solution. Weventure to hope that the account of the reproductive cyclewhich we are now able to present will be found to fulfil, insome measure, expectations iu this latter regard.

We desire to say here that our work has been greatlyfacilitated by the circumstance that some of the phenomenarelating to reproduction in Dasyurus have already beendescribed in greater or less detail (Hill (13, 14), O'Donoghue(20, 21), Sandes (22)).

The breeding season has been divided up into periods for the

11EPH0DUCTIVE CYCLE IN DASYURUS VIVERRINUS. 135

purposes of description, and the terminology employed is, withslight modification, that suggested by Heape (11). As thecestral cycle in Dasyurus differs considerably from that of theEutherian mammal, it has been found necessary to introducetwo new terms, viz. P o s t - c e s t r u s , to designate the periodwhich intervenes between oestrus and ovulation; and Pseudo-p r e g n a n c y , to designate the period which, in the non-pregnant animal, follows ovulation, and in which the changesin the ovary, mammary glands and uteri are essentiallysimilar to those in the pregnant female.

We wish to express our thanks to Mr. F. Pittock, of theZoological Department of this College, for invaluable help inthe preparation of the photomicrographs on Plates 6-8.

MATERIAL.

The information relating to the breeding habits of theanimal was obtained largely from the records mentionedabove. These records relate to 170 females, which fall intotwo classes—pregnant and non-pregnant. Of the non-preg-nant females killed, 13 were prior to ovulatiou, 19 were afterovulation, and in 6 there was no record of ovulation. Ofthe pregnant animals, 37 had less than twenty embryos, 35had more than twenty embryos, in 25 there was no definiterecord of the number of embryos, and 35 were post-partum.Examples of individual records relating to these femaleshave been given previously (20), and further examples aregiven later.

For the purposes of the present paper, the uteri of sixteenfemales were cut in serial section in order to study thehistological changes occurring therein. These uteri were,with two exceptions, from non-pregnant animals both beforeand after ovulation, and were fixed either in picro-corrosive-acetic acid, strong Flemming's fluid, or in Hermann's fluid,all of which gave very good fixation, Flemming's fluid beingparticularly good for the cilia in the uterine glands. Thesections (about 8 ji thick) were stained with hasmatoxylin

136 = J. P. HILL AND CHAS. H. O'DONOGHUE. ••••

and eosin. These sections were compared with those ofpregnant and post-partutn uteri already in the possessionof one of us, whilst we also had access to the numerouspreparations of ovaries and mammary glands of Dasyurus,which formed the basis of the papers of O'Donoghue (20, 21)and Sandes (22) on the corpus luteum and the growth of themammary apparatus. We are thus in the fortunate positionof being able to take into consideration and to correlate thechanges which occur during the cestral and pregnancy cyclesia the several parts of the reproductive and accessory organsmore accurately and in greater detail than, we believe, hasyet been done for any Eutherian mammal.

ANCESTKUS.

The Australian native cat, Dasyurus v i v e r r i n u s , is asmall marsupial somewhat resembling, a civet in externalappearance. It is readily obtainable and fairly easy to keepalive and breed from in captivity.

In the female, the pouch1 is a well-marked structurewhich, during the ancestral period, appears as. a small,somewhat circular depression, situated in the median linetowards the posterior end of the abdomen. It is about 10mm. in diameter by 5 mm. in depth and its boundary isslightly less marked at the anterior end. The interior of thepouch, which in the resting animal is dry and dirty, containsas a rule six teats (vide 20). The teats are arranged inpairs on either side of the middle line and each teat has athickened ledge around its base. The skin over the teat itselfis generally free from the sebaceous glands characteristic ofthe skin as a whole, and which are particularly well marked inthe lining of the pouch. Examination of sections throughthe teat shows that the ledge around its base marks the placewhere the sebaceous glands of the pouch leave off. The teatwith its ledge is situated in a depression, which, in its turn, issurrounded by a raised ridge very nearly circular in shape.

1 c.f. O. Katz, ' Zitschr. wiss.,' Bd. xxxvi, 1882.

REPRODUCTIVE: crci,E IN DASYURUS VIVERBINUS. 137

The ridges around tlie posterior pair of teats are continuousin the middle line, whereas those at the anterior end areseparated from one another by a considerable interval. Thereis also a space between the median walls of the ridges of tliemiddle pair of teats, so that the three pairs are approximatelyarranged in the form of a horse-shoe with the open endsituated anteriorly (see Text-fig. 1, A). The floor of the pouchbetween the ridges surrounding the teats is loose and foldedand frequently falls into a raised fold along the middle line.

The number of teats is subject to slight variation; thus in

TEXT-FIG, 1.

Diagram to show the arrangement of the teats within the pouch,(A) in the normal female with six teats; (B) in the females witheight teats, T. Teat. L. Ledge surrounding teat. i>. Depres-sion, E. Raised ridge. F. Floor of pouch, w. Pouch-wall.

over 170 females examined one pouch contained five teats,two pouches contained seven teats and five contained eightteats1, but all the remainder had the normal number—six.

According to Bresslau (4, p. 672 et seq.) the mammary(nipple) primordia take their origin in Dasyurus as in otherMarsupials from a pair of laterally situated primary anlagen,each consisting of an epidermal thickening and an underlyingarea of condensed cutis. As a rule, three pairs of mammaryprimordia are formed, three from each primary anlage, but

1 Klatsch (quoted by Bresslau (4)) also records the occurrence of 8 teatsin one individual.

138 J. P. HILL AND CHAS. H. o'DONOGHUK

in two pouch-young Bresslau records the presence of four pairs,whilst in a third he found three primordia on one side, fourou the other. In the pouch-young, the mammary primordiaare arranged symmetrically iu two slightly curved longitu-dinal rows, the two primordia of the posterior pair lying-nearest the middle line, those of the anterior pair most remotefrom it, so that the two rows already show the same horse-shoe shaped arrangement as is characteristic of the adult.

Bound each mammary primordium, which has meantimegrown iato the cutis as a knob-shaped epidermal projection,a ring-shaped epidermal thickening arises and this laterhollows out to form a circular groove—the marsupial pocket—bounded by a circular wall. The lateral portions of the wallsof tlie three pockets on each side then join so as to form acontinuous lateral pouch fold and finally the two folds becomecontinuous with each other in front and behind, and so pro-duce the circular wall of the pouch. The occurrence ofsupernumerary nipples (Hypermasty, Hyperthely) in D.Viver r inus is of interest in view of the fact that D. halln-ca tus , according to Oldfield Thomas (28) normally possesseseight teats, that species of Phascologale and Sminthopsis(which genera are regarded by various authorities as beingmore primitive than Dasyurus), have ten, whilst highernumbers still are met wich amongst the Didelphzidas (Perainys-henseli, 17-25). These facts, as Bresslau (4, p. 805-8) haspointed out, would appear to indicate that Dasyurus, like manyother marsupials, has suffered, in the course of phylogeny, areduction in the number of teats within thepouch, which reduc-tion has affected those more anteriorly situated. In the caseof the pouch with only five teats, it seems as if six mammaryrudiments were included in the pouch, but for some reason orother the anterior one on the right side failed to develop.

When the young are born, they are transferred to thepouch, presumably by the mother with the aid of her lips,and, as rhey become permanently fixed to the teats for sometime, the number of teats determines the maximum numberof young that can be reared in one litter.

REPKOnOCTlVE CYCLE IN DASYUEUS VIVJfllRINUS. 139

Dasyurus has only one breeding season1 in the year, whichextends over the winter months, i . e . from the end of May tothe first fortnight in August. - • •

In our records there are two cases of females ovulating inM a y ; one killed on the 21st had eggs which had just enteredt h e uteri , and the other, killed on the 31st, showed corporalutea in a very early stage. The two latest records of p reg-nant females are on the 2nd and 6th of August , and in bothcases.the embryos were in aiKadvanced condition.

The male does not appear to experience an obvious periodof rut , such as Semon (25) describes for the male Phasco-larctos. We have, however, frequently observed tha t thesubst i tut ion of another male for the one previously with afemale frequently resulted in copulation (cf. record, p . 151).The records show tha t copulation may extend intermit tent lyover a period of two to three days. In the majority of cases,however, there is only a single- record of coition, generallylast ing for several hours.- - ,

Then ie thod of copulation in Dasyurus is similar to tha tdescribed, by Selenka for Didelphys (23, p. 105), the maleinounting-on the back of the female and laying hold of theskin of the dorsuin of the neck with his jaws. The penis,when erected, is extremely long and a t tenuated and possessesa markedly bifid glans, its two divisions • doubtless beiuginserted into the corresponding lateral vaginal canals. Therecords show that the spermatozoa may remain alive in theFallopian tubes (where they occur in large bunches in theg land lurnina opening into same) for at least two weeks.

PEO-CESTBUS.

So far as our observations extend, there is no evidence ofany animal ovulat iag twice in the same season, nor is there

1 This term is used in the sense defined by Heape (11), i.e. "to denotetlie whole of that consecutive period during which any male or femalemammal is concerned in the production of young," and does not includethe period of suckling.

140 J. P. HILf, AND CHAS. H. O'DONOGHUE.

any record of a female once served, ngain copulating duringthat season. As a matter of fact, our experience has beenthat a female, once copulation has been effected, will invari-ably fight any male subsequently introduced into the breedingcage. Moreover, study of the ovaries in section has shownthat ovulation, even in females which have not become preg-nant, is not followed by the growth of a second batch offollicles. The cestral cycle in Dasyurus is thus a simple one,and as only one such cycle occurs in the breeding season, theanimal is moncestrous. Our records show that Trichosurus,Phascolarctos (cf: Semon (25) and Caldwell) (5)), Phasco-lomys and probably the Macropods (cf. Caldwell, loc. cit.)breed only once a year, but as to whether they are monces-trous, we have no definite evidence of our own to offer.

Pro-oestrus commences early in June and appears tooccupy a period of time varying from four or five to perhapsas long as ten or twelve days. It is marked by an cedematou&swelling of the lips of the cloaca! aperture, and at the sametime the pouch enlarges somewhat and becomes tumid, whilstits interior sometimes becomes slightly moist. The tumidityis mainly the result of the enlargement of the sebaceous glandsof the pouch area. The sweat-glands also hypertrophy andbecome more coiled. The sebaceous and the sweat-glands soonbecome active; in the former the cells start to undergo auto-lytic disintegration, and in the latter they become granular.The gland lumina increase and become filled by secretion,which, on being discharged, causes the interior of the pouchto become moist and somewhat sticky (vide 20).

Practically no change takes place in the mammary glandsduring this period.

In the ovary, the Graafian follicles gradually enlarge, andas they approach maturity, they form prominences on itssurface.

Uter ine Changes.We have examined sections of the uteri of three females in,

the pro-cestral condition.

REPRODUCTIVE CYCLE IN DASYUUUS VIVEliRINUS. 141

Case 1 (17. vi.}99) (PI. 6, fig. 1).—No record beyonda label in the bottle stating that this female was "probablygetting into heat." The above date shows that she waskilled towards the beginning of the breeding season. Theuteri measured 11 mm. by 7'5 mm. by 7 mm. in diameter.Sections show that, apart from slight increase in size and invascularity, they are practically in the resting condition.The mucosa, averages 1*4 mm. in thickness. The uterineepithelium consists of low columnar cells with close-setnuclei, and has a thickness oE -012 mm.

The uterine glands vai-y in diameter from "032-"048 mm.They are distinctly coiled in the basal half of the mucosa andpossess small but distinct lumina. The connective tissuematrix of the mucosa is for the most part dense and compact.Numbers of leucocytes occur iu it. The blood-vessels extendup between the glands and are already considerably en-larged.

Case 2 (No. 2, 21 . v . '03) (PI. 6, fig. 2).—This femalewas killed on arrival. The pouch appeared reddish aud wasslightly moist. In the ovary occurred practically full-grown ova, with peripheral vesicular nuclei. The mucosaaverages about l -5 mm. in thickness and the uterine epithe-lium, '016 mm. The uterine glands are in a more active con-dition than in Case 1. They are more markedly coiled, arethicker and possess wider luiniua, but the mucosa as a wholeis less vascular.

Case 3 (22 . vi . J01).—No record beyond statement "gettinginto heat." Left uterus, 11 mm. by 11 mm. by 6 mm. Ovawith peripheral vesicular nuclei and not quite full-grown.Graafian follicles small, with retinacula.

The mucosa has a maximum thickness of 2-2 mm. Theuterine epithelium has increased somewhat in thickness, nowmeasuring "02 mm. and is evidently in active growth, mitosesbeing not infrequently met with. The nuclei of its cells arenarrow, elongated and closely packed, and situated at differentlevels.

The basal two-thirds of the uterine glands are markedly

142 J. P. HILL AND CHAS. H. o'DONOGHUE.

coiled and form compact elongate masses, separated from eachother by narrow strands of connective tissue carrying blood-vessels. The lumina of the glands are only occasionallypatent. The gland epithelium is in active growth, mitosesbeing frequent.

The connective tissue, especially in tbe upper third of themucosa, is loose and cedematous in character and has numerousleucocytes dispersed through it. Immediately below theuterine epithelium is a narrow dense zone. The blood-vesselsare enlarging, the mucosa being more vascular than in-Case 2. - •

From the foregoing, it is clear that during pro-oestrus theuteri as a whole enlarge, the mucosa increases in thicknessand becomes more vascnlar, whilst the uterine glands growin length and become markedly coiled. The uterine epitheliumalso thickens.

CESTEUS.

Pro-cestrus is followed by oestrus, i . e . the period of desireon the part of the female, and although in one case thisextended over three days, it appears usually to last only forone or perhaps two clays. As in the higher mammals, it is.only at this time that coition occurs. Selenka (23, p. 104),speaking of Dide lphys marsup ia l i s , states that, "dieBrunst des Weibchens dauerfc jedesmal nur 3-5 Stundeu ! "

Our records show that whilst in one case 1 ovulation mayhave coincided with oestrus, the latter practically invariablyprecedes the former. We have records of five females inwhich copulation had occurred (in three of them, five dayspreviously), whilst ovulation had not yet taken place, asshown by examination of the ovaries. Furthermore, we haverecords showing that unsegmeuted ova were obtained " fromth^, uteri, four (2-celled eggs also present along withivnsegmented), five, six, seven and eight days, after coitus,2-celled eggs sis and seven days after, 4-celled eggs eleven

1 This case is referred to in detail on p. 149.

BJilPUGDUCTIVi; CYCLE IN DASYURUS V1VERRINUS. 1 4 3

and eighteen days after" (Hill, 14, p. 3). It is thus evidentthat ovuhition is not generally coincident with cestrus, butfollows after a longer or shorter interval, amounting in somecases to six or seven or even more days.

Dasyurus thus affords a marked contrast to the Eutheria,in the majority of which cestrus and ovulation are stated tobe generally coincident (cf. Marshall (18), p. 135 et seq.).

During this period the changes in the ovary and pouchstill continue.

Uter ine Changes.

Case 1 (12. vi. '02), PI. 7, fig. 3.—Female killed onarrival. Pouch, reddish, not moist, and only slightly tumid.Cloacal margins distinctly swollen and tumid. Lateralvaginal canals, uteri and Fallopian tubes enlarged, the uterimeasuring 15 mm. x 13 mm. No trace of semen. Ovarieswith prominent Graafian follicles, ova full-grown withperipheral nuclei, shortly before first polar mitosis.

The mucosa has a maximum thickness of 2'5 mm. Theuterine epithelium ("04 mm. in thickness) is formed of verynarrow columnar cells with, for the most part, basally situatednuclei, alternating in arrangement. The uterine, glands(•048--06 mm. in diameter) are markedly convoluted over themajor portion of their extent.. Their epithelial walls areformed of columnar cells, with basally situated nuclei .andwith their inner ends distinctly ciliated, the cilia workingoutwards (PI. 8, fig. 10). Cilia would appear to be presentover the entire extent of the gland lumina, though in spinecases they are less distinct and perhaps absent at the basalextremities of the glands. We. have observed tin's ciliatedcondition of the uterine glands also, in Perameles. It is afeature of interest inasniuch as it affords an, instance of arather uncommon condition, viz. a,glandular epithelium, com-posed of similar cells uniformly, ciliated.. The connective tissue, except immediately.belaw the, u&ei'incvepithelium, where it is dense and very cellular, is in the form

144 J. H. HILL AND CHA.S. H. o'DONOGHUB.

of a very delicate reticulutn, the meshes of which areoccupied by a lightly staining coagulum. Leucocytes arepresent throughout the extent of the mucosa, being especiallyabundant iu its more superficial region, below the uterineepithelium.

The mucosa is not specially vascular.Case 2 (9.vi.;01).—Female killed on arrival. Pouch

with reddish hairs. Not cleaned and not much enlarged.Left uterus 12-5 mm. by 12-5 mm. by 7 mm., right uterus14 mm. by 16-5 mm. by 7 mm. Ovaries with prominentGraafian follicles, ova full grown with peripheral nuclei. Itis noted that this female would not associate with the male.Copulation may therefore have already taken place, but thegenital organs were not examined for sperms.

The uterus agrees closely as regards its histological con-dition with that of the preceding case. The mucosa measuresin thickness from 2'4 to 2-7 mm., and the uterine epithelium•032 mm. The glands generally resemble those of Case 1 andare ciliated.

An important advance is seen in the presence of numerouscapillaries in the connective tissue close below the uterineepithelium, some of which lie in actual contact with the deepsurface of the latter.

Making allowance for individual variation these two casesshow that the growth changes initiated during pro-oestrusare continued without interruption throughout this period.

POST-(ESTRUS.

In the higher mammals, oestrus marks the climax of thepro-osstral changes, and is, according to Heape (11) andMarshall (loc. ci t . ) , the period at which ovulation occurs.It is followed in the non-pregnant female by metoestrum, aperiod during which the various parts of the reproductive andaccessory organs return to a condition of rest.

In Dasyurus, however, we meet with a markedly differentstate of affairs, since we find, as pointed out above, that ovula-tion as a general rule does not take place until some days

REPRODUCTIVE CYCLE IN DASYCTRUS VIVERRINUS. 145

after oestrus. For this period which intervenes betweenoestrus and ovulation, we propose the term " post-oestrus."

During this post-oestral period the changes initiated duringpro-oestrus are continued, and are perhaps most marked iu theovary. The G-raafian follicles continue to enlarge. The ovaundergo the first meiotic division, resulting in the separa-tion of the first polar body, and the spindle for the seconddivision ia laid down. Out- records yield instances of twofemales, both killed five days after copulation, in one of whichthe first polar spindle was established, whilst in the other, thefirst polar body was already separated and the second polarspindle formed. These ovarian changes culminate iu therupture of the follicle and the discharge of the ovum.

The tumidity of the cloacal lips gradually disappears.

U te r ine Changes .

Case 1 (No. 8, 23 . vii.'02) (PI. 6, fig. 4).—This female•came to hand on July 11th. On the 18th the cloacal marginswere greatly swollen, and copulation, extending over twohours, took place. Five days after, i . e . on the 23rd, shewas killed, the pouch being small and only slightly tumid.The uteri were pale in colourand somewhat enlarged, measuring16"5 mm. by 12 mm- by 5'5 mm. Examination of the ovaryrevealed the presence of ripe follicles containing ova in whichthe first polar spindle was already formed.

The mucosa has a maximum thickness of 2*4 mm. and theuterine epithelium, '032-"036 mm. The uterine glands haveincreased somewhat in diameter, and some of them aremarkedly enlarged, especially towards their basal ends.Where such enlargement has occurred the gland epitheliumhas become reduced in thickness and appears of the lowcolumnar type. Iu a few of the glands the lumen is occupiedby a shrunken coagulum. The gland epithelium is ciliated.

The connective tissue differs from that of oestral Case 1in being much more cellular and compact. Leucocytes arespecially abundant ia the superficial portion of the mucosa,

VOL. 5 9 , PART 1. NEW SERIES. 1 0

146 J. P. HrLL AND OHA.S. H. O'DO.NOGHUE.

Capillaries are present here and there below the uterine epi-thelium, but are not specially abundant.

Case 2 (No. 14, 26 .vii .'02).—The following is our note-book record of this female :

24. vi. '02.—Eeceived.4 . vii. '02.—Cloacal margins commencing to swell.

11. vii .'02.—Cloacal margins still swollen.2 1 . vii. '02.—Copulation.24. vii. '02.—Pouch slightly tumid, not moist.26. vii.'02.—Pouch slightly tumid; killed, i . e . five days

after copulation.Examination of the ovaries revealed the presence of mature

follicles containing ripe ova in which the first polar body wasalready separated and the second polar spindle established.

The uiucosa has a maximum thickness of 2'4 mm. The-uterine epithelium has made little or no progress, and appearsas a low columnar epithelium ('02 mm. in thickness) withclose-set ovalish nuclei. Here and there, between the ordinaryepithelial cells, there occur single very narrow cells withdarkly staining cytoplasm and compressed deeply stainingnuclei.

The uterine glands (averaging '048 mm. in diameter) donot differ essentially from those oE oestral Case 2, and arerather less advanced than those of the preceding uterus.They are well convoluted and lined by the usual ciliatedcolumnar epithelium. Their lumina are for the most partdistinct but small.

The connective tissue is more oedematous than that of thepreceding case, and contains in places much coagulum.Numerous leucocytes are present in it. The mucosa is, onthe whole, more vascular than that of Case 1, though the-superficial capillaries are not yet greatly developed.• So far as can be judged from these two cases, it wouldappear that during this post-cestral period, no very markedadvance is made by the uterine mucosa, the uteri being, on.the whole, very similar to those of the cestral period.

REPRODUCTIVE CYCLE IN DASYHRUS VIVEEBINCJS. 147

Ovulat ion.

It has bsen pointed out by Ancel and Bouin (1) that theEutherian mammals may be divided into two classes accordingto the conditions under which they ovulate. In the firstgroup ovulafcion is spontaneous—i. e. it takes place whetherthere has been coition or not (this condition appears to bethe more general); whilst in the second group ovulation isnon-spontaneous—i. e. in the normal course of events, copu-lation is necessary to provoke it.

In Dasyurus, ovulation is spontaneous and quite independentof copulation. We have records of nine females whichovulated after being under observation for periods varyingfrom two to thirty-seven days, during which no copulationwas seen to take place. Four of these females were underobservation for twenty days and upwards, whilst in three ofthem, the unfertilised ova were found in the uteri. On theother hand, as pointed out above, we have records of fiveanimals in which copulatiou had taken place but ovulationhad not, three of these females having been killed five daysafter copulation. This evidence shows conclusively thatbyulation and copulation are perfectly independent one ofthe other, and that the former is in no way provoked by thelatter, as is the case, for example, in the rabbit.

A further point in connection with ovulation is worthy ofattention, although it has already beeu noted elsewhere(Hill (14)), and that is the remarkable number of eggs dis-charged from the ovary at each ovulation. In one casetwenty-eight eggs, in two others thirty eggs, and in yetanother thirty-five vesicles were obtained from the two uteri."There can be no doubt that Dasyurus, like various othermarsupials—e. g. Perameles, Macropus, etc.—has suffered aprogressive reduction in the number of young reared, but, evenmaking due allowance for that, the excess in production overrequirements-would still be remarkable enough " 1 (Hill (14) ).

1 Examination of the ovaries of D. maculatus with full-grownfollicles shows that the same excessive production of ova holds true also-for this species.

148 J. H. HILL AND CHAS. H. o'DONOGHUE.

Ovulation may be succeeded by one of two states. On theone hand, if the ova be fertilised, there follows a period ofpregnancy, which in its turn is succeeded by a nursingperiod. On the other hand, in the absence of copulation orfertilisation, there follows an interval in which the variousorgans undergo a series of marked alterations, essentially ofthe same uature as those normally undergone in the pregnantand post-partum female. To this pei-iod, therefore, we pro-pose to give the name of " p s e u d o - p r e g n a n c y . "

PREGNANCY.

If fertilisation is effected, ovulation is normally succeededby pregnancy. The ova are fertilised in the upper parts ofthe Fallopian tubes, and the second polar body is there givenoff. They pass down the tubes, apparently with considerablerapidity, into the uteri, where cleavage begins (Hill (14) ).

We have already directed attention to the fact " that thereis no correlation between the number of ova shed duringovnlation and the accommodation available in the pouch(Hill (14)), the number of ova shed at one period being, as arule, far in excess of the normal number of teats (sis). Ourrecords show that out of seventy-two pregnant females inwhich there is a definite record of the number of eggs orembryos, thirty-five had more than twenty, and the remainingthirty-seven less than twenty. Moreover, whilst amongst thethirty-five one had as many as thirty-five vesicles, in onlythree of the remaining thirty-seven were there less than sixembryos, and in one of these only one uterus was pregnant.Although a proportion of the eggs may for one reason orother fail to develop normally, it would appear to happengenerally that a larger number of young are born than canpossibly survive, the pouch accommodation being strictlylimited. Our records afford two specific instances of femaleswhich were examined shortly after parturition. In one ofthese, eighteen young were found, of which eleven occurredadhering to the hairs round and directly below the opening ofthe pouch, six were attached one to each nipple, and one

REPRODUCTIVE CYCLE IN DASYURUS VIVERRINUS. 149

occurred free in the pouch alongside an attached one. In theother, ten young were found, six attached, three free in thepouchj and one outside considerably shrivelled.

Gestat ion Per iod.

Attention has already been called to the fact that ovulationis entirely independent of copulation, a fact which renders itextremely difficult, if not impossible, to determine accuratelythe length of the gestation period. One of us has alreadyrecorded that " the shortest period observed between coitusand the birth of the young was a little over eight days,"hence it was concluded " that the time of gestation does notexceed this period" (14). Selenka, it may be noted, givesthe gestation period of Didelphys as barely eight days (23).As regards our eight-day record, we think it is desirable togive further details. Copulation was observed on June7th, about ten a.m. It lasted a comparatively short time.The female was killed on June 15th at six p.m. A singleyoung one (recently born, and measuring 7"5 mm. greatestlength in the fresh state) was found in the pouch, which wasgreatly enlarged and possessed prominent sebaceous glands.Our notebook record continues, " assuming that copulationhad not taken place before receipt of this female, we can putdown the gestation period as a little over eight days." Thereare two points in this record which should be noted : first, thebrevity of the copulation, and second, the question of copula-tion previous to receipt. The recollection of one of usregarding this particular female is that she was brought in bythe collector, placed at once with the male, a short copula-tion, or an attempt at the same, ensuing almost immediately.

In view of the impossibility of entirely excluding theoccurrence of copulation previously, we cannot regard theforegoing evidence as absolutely conclusive as to the length ofthe gestation period. At the same time it should not beforgotten that a female once served will not under normalcircumstances again copulate. It is of course quite possible

150 J. P. HILL AND CHAS. H. o'DONOGHUE.

that the circumstances in this particular case were not normal,since this female was no doubt in a cowed and frightenedcondition when placed in the cage with the male.

We have another record of a female1 in which the youngwere born sixteen days after copulation. This particularfemale was received on June 20th. Copulation occurredon the 24th, and parturition on July lOfch. This recordwe regard as perfectly trustworthy. The question as to whenovulation occurred, however, cannot be answered with anydegree of accuracy. We have already stated that ovulationoccurs at a variable period after oestrus. In our records wefind that the shortest time intervening between copulationaud the finding of unsegmeuted eggs in the uteri is fourdays (in this particular case the eggs were in the unsegmentedand 2-celled stages), and the longest time, eight days.Between these limits we have l'ecords of the following: fivedays after, no ovulation (three cases) ; five days after, un-segmented ova (one case) ; five days after, unsegmented aud2-celled eggs (one case); five days after, 4-celled eggs(one case) ; six days after, unsegmented ova (one case) ; sixdays after, 2-celled eggs (one case) ; seven days after,unsegmented ova (one case).

The average interval between copulation and ovulationwould thus appear to vary round about five or six days. If,now, we subtract five from the sixteen-day record givenabove, we have a gestation period of about eleven days.From the evidence we think we are justified in stating thatthe gestation period in Dasyurus is not less than eight, anddoes not exceed fourteen days.

Genera l Changes .

The ovarian changes following ovulation, i . e . those result-ing in the formation of the corpora lutea, have been fully

1 This female is the one previously referred to (ante, p. 148) as havinggiven birth to eighteen young.

REPRODUCTIVE CYCLE IN DASYURUS VIVERRINUS. 151

dealt with by Sandes (22). It need only here be mentionedthat the corpora lutea essentially resemble those of Butheriain their mode of development and structure. Sandes (p. 380)states that "the corpus luteum forms quickly, within threedays [after ovulation], and persists [not only throughoutpregnancy but] during the greater part of the time that theanimal is lactating, ultimately disappearing when the younganimal is capable of leading an independent existence." Thechanges iu the pouch, including its sebaceous, sweat, andmammary glands have been described elsewhere (O'Donoghue(20)), whilst a preliminary account of the arrangement ofthe foetal membranes, the placentation, and the mode ofparturition, has also been published (Hill (13)).

Early P regnan t Ute r i .

Case 1 (No. 15, 19 . vii. '01) (PI. 6, fig. 5).—We give herefor purposes of comparison a brief account of the uterus of apregnant female with ova at the stage before the separationof the yolk-body and shortly after entering the uteri.

The record of this female is a very complete one, extending• over a period of forty-one days. We reproduce it here :

8 . vi. '01.—Resting; so on to 6 . vii. '01.9 . vii . '01.—Oloacal margins very slightly swollen.10 . vii .'01.—Fresh male in; no copulation.11. vii. '02.—Cloacal margins swollen, but not greatly.13. vii.'02.—Cloacal margins swollen. Fresh male iu ;

copulation.17. vii.'02.—Pouch hardly altered; cloacal swelling still

present. ... .18 . vii. '02.—Pouch very slightly tumid, very dirly.19 . vii • '02.—Pouch distinctly tumid, not cleaned.Killed, i .e . six days after copulation.The uteri measured 1'3 cm. by 1-3 cm. by -7 cm., the left

containing ten ova and the right fifteen.•The m-ucosa has a maximum thickness of 2-l mm. The

uterine epithelium (1016-'018 mm. thick) is little advanced,

152 J. P. HILL AND CHAS. H. O'DONOGHUK.

appearing as a low columnar epithelium with close-set nuclei,plump and rich in chromatin. The uterine glands (04-•05 mm. in diameter) are well convoluted, luminated through-out, and ciliated. The connective tissue is markedly oede-matous in the superficial half of the mucosa, and particularlyrich in leucocytes; in the deeper part of the same, betweenthe basal coils of the glands, it is more cellular. Below theuterine epithelium there is the usual compact zone of con-nective tissue, which is rich in leucocytes and fairly vascular.

Case 2 (No. 13,25 . vii. '02).—Female killed four days afterfinal copulation, with 1 and 2-celled ova and abnomialsin the uteri. The uterus is essentially similar to that' ofCase ] . Mitoses in the uterine and glandular epithelium arenot infrequent.

THE NURSING PERIOD.

Parturition in mammals is followed by a period termed byHeape (11) the " nursing period," during which the young arenourished by the milk secreted in the mammary glands ofthe mother. As is known, there is a noteworthy differencein the way in which the milk is obtained by the young in theEutheria and Metatheria. In the former the young are per-fectly free and seek the teat of the mother,from which they suckthe milk. The new-born marsupial, on the other hand, whichis brought forth in a much less advanced condition than theEutheriaD, is transferred to the pouch, presumably by themother with the aid of her lips, and becoming fast to a teat,the milk is pumped down its throat.

The nursing period in the marsupial falls into two distinctphases :

(1) A pe r iod of fixation, during which the lips of theyoung one are fused in such a manner that it is firmly attachedto the teat, which it cannot leave. During this period themilk is said to be forced periodically into its mouth by thecontraction of the muscles of the pouch area of the parent.This period in Dasyurus lasts from seven to eight weeks (13).

REPRODUCTIVE CYCLE IN DASYUEL'S V1VJEREJNUS. 153

(2) A f ree p e r i o d , during which the lips of the younganimal are no longer fused, so that it can leave the tent atwill, but is still entirely dependent on its mother for food. InDasyurns, this period extends over eight or nine weeks. Thetotal time occupied by these two periods, which togetherform the true nursing period, is about four months in Dasy-urns (13).

After this period, the young move about freely away fromthe mother, and begin to eat, although they still may makeoccasional use of the pouch.

This marks the conclusion of the reproductive cycle, andthe various organs now return to an anoestrous condition, orstate of rest, until the following year.

PSEUDO-PBEGNANCY.

Ovulation is followed by the formation of corpora lutea.According to Ancel and Bouin (1), in animals in which ovu-lation is spontaneous, two distinct kinds of corpora lutea areformed; (a) iu the pregnant animal, a gestative corpusluteam (corpus luteum verutn), which persists for a long time,and (b), iu the non-pregnant animal, a periodic corpusluteum (corpus luteum spurium), which is not so fullydeveloped as the former, and has only a transitory existence-No such distinction is possible in Dasyurus, for the structureof the corpus luteum is identical, whether pregnancy followsovulation or not, and there is evidence to show that iu thenon-pregnant animal the corpns luteum persists for a consid-erable time, some weeks at least, and even then shows no signof degeneration (vide 21).

The pouch continues to enlarge after ovulatiou, and thesweat and sebaceous glands of the pouch area reach a stateof development and activity comparable to tliat attained inthe pregnant animal. In some cases, it was observed that thefemale started to clean out the pouch in the same way as doesthe pregnaut female, in preparation for the reception of theyoung.

154 J. P. HILL AND CHAS. H. o'DONOGHUE.

The most striking changes during this period, however,occur in the mammary glands. These changes have alreadybeen fully described (20), so that here it is only necessary toemphasise the fact that the glands hypertrophy in preciselythe same way as do those of the pregnant animal, and ulti-mately reach a state of development, which is at least asadvanced as that in a female thirty-six hours after the birthof the young.

Uter ine Changes.

C a s e l ( 3 1 . v . '01).—This female died in captivity the dayafter its arrival. The uteri were slightly enlarged, but no ovawere found. Examination of the ovaries reveals the presenceof corpora lutea in the last stages of growth, there beingpresent in their central region small spaces not yet filledeither by lutein cells or by connective tissue. Ovulation musthave occurred some days previously.

The uterine mucosa measures 2—3 mm. in maximum thick-ness. The uterine epithelium ('027 mm. thick) is formed ofnarrow columnar cells with closely packed small nuclei, ovalishor rounded in form, occupying the mid-region of the layer.The utei'ine glands appear crowded together and are markedlyconvoluted. The connective tissue is compact below theepithelium and fairly so between the glands. The mucosa isnot specially vascular.

This case is o£ importance as showing the condition ofthe uterus in a female some time after ovulation and before theonset of the degenerative and regenerative changes in themucosa.

Our next two cases are of special iuterest as showing stagesin these degenerative and regenerative changes in the non-piegnant uterus after ovulation, changes which we areconvinced are the homologuesof those seen in the normal post-partum uterus.

Of Cases 2 and 3 we have no further record beyond thestatement that the uteri were opened but nothing was found.

REPRODUCTIVE CYCLE IN DASYURUS VIVERRINUS. 155

Case 2 (vii. 06) PI. 7, fig. 6. Corpora lutea are presentin the ovaries. They are at a stage towards the eud of thegrowth period, and in their degree of development are inter-mediate between those of Case 1 and Case 3.

The mucosa has a maximum thickness of 2'9 mm. Theuterine epithelium appears as a layer of quite low cubical cells{•006—"009 mm. in thickness), which essentially resembles thatof a uterus seven days post-partum. Vacuolar spaces occurhere and there in the epithelium, but the nuclei on the wholeare plump and rich in chromatin and show no obvious signsof degeneration. It is therefore probable that the uterineepithelium has already been reconstituted.

The uterine glands are in a very interesting condition, theappearances presented strikingly recalling those seen in theglands of post-partum uteri. Over the major portion of thethickened area of the mucosa, the basal portions of the glandsare markedly couvoluted and greatly hypertrophied but showno signs of degenerative change. They average in diameter•072 mm. (varying from '056—-1 mm.), and are lined by anepithelium composed of very narrow, high columnar cells(•024—-04 mm. in height), with small basally situated nuclei.The gland lumina are distinct, but small, and in many casescontain a homogeneous secretion which stains deeply withesoin. The convolutions of the glauds are very compactlyarranged, with little or no connective tissue between them.

Comparison of these hypertrophied glands with those of apregnant uterus with blastocysts 1*5 mm. in diameter showsthat whilst the gland convolutions in the latter are not socompactly arranged, the glands themselves are slightly thicker(08—"096 mm. in diameter, epithelium "032 mm.) and the

gland lumina are much larger. .In the more superficial region of the thickened area of the

mucosa, as well as round its periphery, the glands and theportions of them situated therein present a very differentappearance. They are much less convoluted and much lesscompactly arranged. Moreover, instead of possessing thickglandular walls and small lumina, they are lined by an

156 J. P. HILL AND OHAS. H. O'DONOGHUE.

epithelium of the quite low .cubical type ('006—-015 nun. inthickness) and possess large and distinct lumina. Theselatter are not empty, but contain either a homogeneouscoagulum in which cells may be embedded, or, as is more of tenthe case, they are occupied by a more or less compact mass ofcells (PL 7, fig. 9). The cells possess large, rounded orpolygODal cell-bodies which stain deeply with eosin, and nucleiwhich are occasionally found in division and which are quitesimilar to the nuclei of the surrounding epithelial cells.

The transition between these altered portions of the glandsand the more deeply situated unaltered or rather liypertrophiedparts is somewhat abrupt, so that it is difficult to obtain verydefinite evidence as to how the intra-luminar cells originate.We are of opinion, however, that they are derived from thehypertrophied gland epithelium by a kind of desquamationprocess, and that they increase in size after their separation.If that view is correct, then we must regard the epitheliumof the thin-walled portions of the glands as being in pro-cess of regeneration. In this connection, it is worthy ofnote that mitoses are not rare in the epithelium, and that thesuperficial portions of the glands close below the openingsinto the uterine cavity are ciliated.

The connective tissue is no longer cedematous, but appearsas a compact, richly cellular tissue. Leucocytes are notspecially abundant, but are most numerous in the moresuperficial region of the rnucosa, where there are also presentnumbers of capillaries of varying size. It is worthy of notethat there are no blood extravasations.

The uteri in this Case, with the uterine epithelium and muchof the gland epithelium regenerated, may therefore bedescribed as well on the way towards the resting condition.The uterine epithelium resembles that of the seven days'post-partum uterus, the glands, those of the two days' post-partum uterus.

Case 3 (1900) (PI. 8, fig. 7).—The ovaries show corporalutea at the end of the growth period, just older than those ofCase 2.

REPRODUCTIVE CYCLE IN DASTtJRUS VIVERRINPS. 157

The mucosa has a maximum thickness of about 3'5 mm.and presents a markedly folded surface. The uterineepithelium appears as an irregular folded layer of lowcolumnar cells and shows evident signs of degenerativechanges, e.g. irregularities in contour both of its cells andnuclei, occurrence of vacuoles in the cytoplasm, presence ofdegenerated nuclei and of diffused darkly staining (chroma-tinic ?) granules. Moreover, blood extravasations from thesuperficial capillaries would appear to have taken placethrough clefts formed by the breaking down of the epithelialcells. These extravasations appear to have been slight, butare undoubted. Besides corpuscles, there are present inthem, cell remnants and small, darkly staining granules. Inthis female, accordingly, the uterine epithelium, unlike thepreceding case, has not yet been reconstituted.

The uterine glands are in essentially the same condition asthose of Case 2. In the deep portion of the central region ofthe thickened area of the mucosa, the parts of the glandstherein situated appear for the most part closely packed andhypertrophied. The gland epithelium is formed by high,columnai1, pale-staining cells, often vacuolated, and with smallbasally situated nuclei. The gland lumina are distinct andin some cases contain a staining coagulum.

Elsewhere the glands are in process of being reconstituted.They are lined by a low flattened to cubical epithelium andtheir lumina are occupied either by coagulum with embeddedcells or by cellular masses, which differ from those of Case 2only in that the cells are in course of degeneration.

The connective tissue differs strikingly from that of Case 2,being in the form of a fine reticular tissue, oedematous andextremely rich in leucocytes.

Numerous enlarged blood-vessels are present in the mucosa,whilst in actual contact with the under surface of the uterineepithelium there are numbers of fine capillaries such as occurnormally in this position in the late pregnant uterus.

Case 4 (31 . vii. '99) (PI. 7, fig. 8.—The record is asfollows :

158. J. P. HILL AND OHA.S. H. O'DONOOHUE.

14 . vii. '99.—Female brought in.15 . vii. '99.—Put with, male ; no record of copulation.17 . vii . '99.—Pouch large, dirty.20 . vii. '99.—Pouch large, fairly clean, teats red, sebaceous

glands just appearing.22 . vii. '99 \24. vii . '99 I Pouch relaxed, but nob quite clean, glands26 . vii . '99 j small.28 . vii. '99 J31 . vii. '99.—Pouch quite relaxed, moist, glands fairly well

marked; apparently pregnant.Killed, but nothing found in the uteri, which were enlarged.

The ovaries were sectioned and show full-grown corpora lutea,somewhat older than those of Case 3.

The mucosa is thickened and its surface is thrown intowell-marked folds. The uterine epithelium is very similar tothat of Case 3, and is for the most part in process of degenera-tion. In places it appears as a practically unaltered layer ofcubical cells, but over most of its extent, it exhibits evidentsigns of degenerative change. It is very irregular both incontour and thickness and is invaded by leucocytes. More-over, cells may be seen in process of desquamation from itsouter surface, and in places, it appears actually to have brokendown, since extravasated blood, in which occur degeneratecell-products, is present in small quantities in the uterinecavity close to the surface of the epithelium.

The uterine glands have been reconstituted throughout theirextent. They are lined by a quite low epithelium, and theirlumina are occupied more or less completely by cellularmasses (containing leucocytes), in a more or less advancedstage of degeneration.

The connective tissue is very cellular and is invaded bylarge numbers of leucocytes.

Making allowance for individual variations due to ageor other causes, the foregoing Cases afford evidence of theoccurrence of progressive, regressive and regenerative changesin the uterine mucosa, during this period of pseudo-pregnancy.

REPRODUCTIVE OYOLE IN DASYURUS ViVERRIKTUS. 1 5 9

(a) Progressive Changes.—The mucosa as a wholeundergoes marked thickening, its vessels enlarge, and in par-ticular a rich plexus of capillaries is established immediatelybelow the uterine epithelium. The uterine glands hypertrophy,their epithelial cells .assume an elongate columnar form andsecrete actively.

These changes are identical with those seen in normally preg-nant uteri. The only noteworthy difference in the progressivealterations in the pseudo- and normal pregnant uterusconcerns the uterine epithelium, the cells of which in theformer fail to become transformed into the high columnarelements characteristic of the latter.

(b) Eegressive Changes.—The uterine epithelium doesnot appear to be shed as a whole, but undergoes partial de-generation and desquamation, and it is noteworthy thatthese degenerative changes may so effect the immediatelyunderlying capillaries as to lead to their rupture and the con-sequent appearance of extravasated blood in small quantitiesin the uterine cavity.

We have so far not observed such extravasation in normalpost-partum uteri, though it is possible that this may occa-sionally occur. On the other hand, the very marked extravasa-tions into the connective tissue which are found in post-partumuteri are not met with in the pseudo-pregnant cases.

The uterine gland epithelium also appears to undergo aprocess of desquamation—at all events cells are shed from itinto the gland lumina, where they form cellular masses whicheventually degenerate. Comparison of our preparations ofpseudo-pregnant uteri with those of normal post-partum uteridemonstrates beyond all possibility of doubt that the regres-sive changes in the glands are identical in the two. In post-partum uteri, we find the same reduction of the gland epi-thelium from the high columnar to the low cubical type takingplace, and accompanied, as in the pseudo-pregnant uterus, bythe appearance in the gland lumina, of masses of cells whicheventually degenerate.

(c) Regenerative Changes.—The uterine epithelium

160 J. P. H[Lr, AND CHAS. H. o'DONOGHDB.

is reconstituted, and the low cubical epithelium of the uterineglands persists like that of the post-partum uterus to formthe lining of the resting glands. The connective tissue re-assumes its more compact form, and the inucosa as a wholeundergoes marked decrease in thickness and returns to thecondition of rest.

It is difficult, indeed, impossible, to nscertain the exactduration of this period of pseudo-pregnancy, but so far as wecan judge with the aid of our i-ecords, it probably extends overabout two weeks. We have, for example, a very full record ofa female, apparently in heat about Juue 28th and killed onJuly loth, i . e . seventeen days after heat, and of anotherfemale in heat on June 9th (copulation on June 9th to 12th)and killed on June 30th, i . e . twenty-one days after heat. Inboth, nothing was found ia the uteri. In both, the ovariesshowed old corpora lutea, the pouch was greatly enlarged themammary and other glands being well-developed, whilst theuteri were extremely large and vascular (measuring in thefirst-mentioned female by 4 by 1 cm. and in the second 39by 4'1 by T6 cm. in diameter).

METCESTRUS.

After the changes described in the preceding section havetaken place, the whole of the reproductive organs graduallyreturn to a state of rest. This metoestral period correspondsfunctionally to the similarly named one in the Eutheria, onlythere is the striking difference that, whereas in the unimpreg-nated Eutherian mamma], metcestrum follows immediatelyafter ovulation, in the unimpregated marsupial, the two areseparated by post-oestrus and pseudo-pregnancy, the twoperiods together occupying at least a fortnight, and probablylonger.

SUMMARY.

Dasyurus is monoestrous and has one breeding season ayear, which begins at the end of May or early in June and

REPRODUCTIVE CYCLE IN DASYURCJS VEVERRINPS. 1 6 1

lasts into the first fortnight in August (i.e. it extends overthe winter months).

The male does not appear to experience a marked ruttingseason.

Copulation is similar to that of Didelphys (Selenka), andthe sperms can remain alive in the Fallopian tubes for atleast two weeks.

Ancetrus.—The ancestral period lasts more than half theyear.

Pro-cestrus.—Pro-oestrus appears to extend over a vary-ing period of from four to twelve days.

During this time, the lips of the cloaca become swollen, thepouch enlarges somewhat and becomes slightly tumid andmoist, and the Graafian follicles increase in size and becomevesicular. The uterine mucosa increases in thickness andbecomes more vascular, its glands lengthen and become con-voluted and the uterine epithelium also tends to thicken.

CEstrus.—CEstrus lasts usually for one or two days and isthe period during which copulation occurs.

The changes already initiated during pro-oestrus in thevarious parts of the reproductive system are continued with-out interruption.

Post-oestrus.—Post-oestrus, which term we employ todesignate the period following cestrus and terminated byovulation, occupies as a rule about five or six days.

The tumidity of the cloacal lips disappears, but the changesin the pouch and uterus still continue, not, however, veryactively.

In the ovary (1) the ova give off the first polar bodyand the spindle for the second meiotic division is formed.

(2) The follicles attain maturity and ultimately rupture,setting free the ova.

Ovulation.—Ovulation marks the end of this period andoccurs generally about five or six days after cestrus. It isspontaneous and independent of copulation and is remarkablebecause of the large number of ova liberated. Ovulation issucceeded (a) by pregnancy or (b) by pseudo-pregnancy.

VOL. 5 9 , PART 1.—NEW SERIES. 11

162 J. P. HILL AND OHAS. H. o'DOHOGHUE.

Pregnancy.—Fertilisation is effected in the upper partof the Fallopian tube and the second polar body is theregiven off.

As a rule more young are born than can possibly surviveowing to the limited accommodation in the pouch.

The g e s t a t i o n pe r iod is not less than eight and notmore than fourteen days, but the interval between copulationand birth is usually considerably longer.

Corpora lutea are formed and persist throughout the greaterpart of the time that the animal is lactating.

N u r s i n g Period.—This period may be divided into twophases.

(1) Per iod of F ixa t ion .—A period, lasting for seven oreight weeks, during which the young are firmly attached tothe teats.

(2) F r e e Per iod .—A period of eight or nine weeks whenthe young are free in the pouch but dependent on the motherfor food.

After this time the various organs gradually return to astate of rest.

P seudo-p regnancy .—We have applied this term to theperiod following ovulation iu cases where the ova have failedto develop, because of the occurrence in it of a series ofchanges in the reproductive organs essentially identical withthose met with in pregnant females.

Corpora lutea are formed in the ovaries which are identicalwith those in the pregnant female.

The pouch enlarges, and its sweat and sebaceous glandsreach a state of hypertrophy and functional activity com-parable to that in the pregnant female.

The mammary glands also enlarge and reach a state of•development equal to that in a female thirty-six hours afterparturition.

The uteri enlarge and become vascular, often to a marked•degree.

The uterine mucosa undergoes a series of changes, pro-gressive, regressive and regenerative.

THE GESTRAL C y « , E IN THE

General symptoms.

ANCESTRTTS . . . Rest.PJBO-OBSTRUS . . Swelling of cloacal lips.

(ESTRTFSPOST-CESTBtTS

PSEUDO-PKEGNANCr .

METCESTRUS

Pouch commences to be-come tumid.

Period of desire.Pouch tumid and slightly

moist, growth of sweat andsebaceous glands of poucharea. Regression of cloacal

Furtherenlargementof pouchaccompanied by growth ofmammaiy glands as in preg-nancy.

Return to rest.

N~ON-1>EEGNANT FEMALE D A S Y U K U S .

Uterine changes.

Rest,Commencement of con-

s t r u c t i v e s t age . Hyper-trophy of inucosa with in-crease of vascularity. Pro-liferation in uterine epithe-lium and glands.

Continuance of above.Continuance of above but

not very actively.

Continuance of above, (a)Progressive changes inuterine glands especially.(b) Regressive (destruc-tive) changes, e.g. de-generation and desquania-tion of uterine and glandepithelium. Extravasationof blood, (c) Regenera-tive (repair) changes.Recuperation of tnucosa.

Return to rest.

Ovarian changes.

Rest.Growth of folh'cle and con-

tained ovum.

Continuance of above.Follicles attain maturity.

Separation of first polarbody and formation ofsecond polar spindle. Ovu-

Fonnation and growth ofcorpus luteum as in preg-nancy.

Return to rest.

ANOSSTKTTS.—As Dasyurus is moncestrous, metcestrus is followed by anoestrus, which lasts until the following year.

T H E CESTRAL CYCLE IN THE

ANCESTETTSPBO-CESTEUS

OESTRUS

METCESTRTTS

General symptoms.

Rest.Swelling and flushing of

vagina. General symptomstermed by breeders "comingin heat."

Appearance of menstrual dis-charge if present.

Period of desire. Vaginastill swollen and flushed.

Return to rest.

NON-PREGNANT FEMALE EDTHERIAN.

Uterine changes.

Rest.Constructive s tage :

Swelling of mucosa. Pro-liferation in uterine epithe-lium and glands. Hyper-semia and congestion, Des-timetive stage: Increasedhypersemia. Breakdown ofvessels and formation ofblood lacunEe.

Rupture of lacunae, degenera-tion of uterine and glandularepifchelia. Fomiation ofmenstrual clot if present.

Stage of repair : Recupe-ration of various uterineelements.

Return to rest.

Ovarian changes.

Rest.Growth of Graafian follicle.

Maturation of ovum.

Discharge of ova (in animalswith spontaneous ovulation).

Formation of corpus luteumspurium (in animals withspontaneous ovulation).

REPRODUCTIVE CYCLK IN DASYURUS V1VERRINUS. 165

Metoestrus.—This is an indefinite period during whichthe whole of the reproductive organs return to a state of rest.

CONCLUSIONS.

Comparison of the (Estral Cycles in the Metatheriaand Eufcheria.

A comparison of the tables of the cestral cycle in the non-pregnant Dasyurus and in the non-pregnant Eutherian (pp.163-164); following the schemes of Heape (11) and Marshall(18), discloses certain rather striking and important differencesbetween them. These differences are seen at a glance in the

SCHEME OF REPRODUCTIVE CYCLE IN—

(1) DASYURUS.

Anoestras.

Pro-cestvus.

CEstrus.

Post-cestrus (ovulation).

(2) EUTHERIAN.

Ancestrus.

Pro-oestrus (uterine degeneration).

ICEstrus

(ovulation).

Dicestrus.1

Metoestrus.

Pregnancy. Pseudo-pregnancy Pregnancy,(uterine degeneration). I

Nursing period. Metoestrus. Nursing period.! I I

Anoestrus. Ancestrus.1 This period only occurs in polycestrous animals.

scheme which we have drawn up of the reproductive cyclein the Marsupial (as represented by Dasyurus), and what maybe taken to be the usual cycle in the Eutherian mammal asset forth by Heape and Marshall. They concern more espe-cially the time-relations of the degenerative uterine changes

166 J. P. HILL AND OfUS. H. O'DONOGHUE.

and of ovulation to the cycle as a whole. Thus, taking cestrusas an easily recognisable and obviously equivalent event iuthe two cycles, we see that, whilst in the Marsupial, ovulationonly occurs at an interval of some days after oestrus, inthe Eutherian, ovulation either coincides with, or followsimmediately after cestrus. We have emphasised this importantdifference by applying the term " post-oestrus " to the periodintervening between oestrus and ovulation.

Furthermore, there is a very striking discrepancy in thetime'of occurrence of the degenerative changes in the uterinemucosa in the two. In the marsupial, tliese succeed ovulation ;in the Eutherian they not only precede ovulation, but alsocestrus. This, we regard as the most striking difference inthe two cycles.

In the preceding pages we have produced evidencedemonstrative o£ the essential similarity in Dasyurus of thesedegenerative changes in the pseudo-pregnant and post-partumuteri. Indeed, comparison of the entire series of changes inthe organs directly concerned with and related to reproductionin females which have ovulated and have failed to becomepregnant, and which we have already described (ante, pp. 21 -28), with those seen in the same parts of the pregnant female,entirely justifies us in recognising the occurrence of what wehave termed a period of pseudo-pregnancy in the cestralcycle of tlie marsupial and in affirming that that pei-iod repre-sents the phase of true pregnancy.

That being so, the question arises, What part of the Euthe-rian cycle corresponds to pseudo-pregnancy in the marsupial ?There can be little doubt but that the uterine degenerativechanges seen during the pseudo-pregnancy period in themarsupial are equivalent to those which take place in theEutherian uterus during pro-oestrus, a conclusion which wethink, in view of the evidence herein presented, will meetwith general acceptunce. •

These pro-cestral changes in the Eutherian uterus con-dition the appearance in some members of the sub-class(e.g. Primates) of a sauguineous discharge—the menstrual

REPRODUCTIVE CYCLE IN DASYURUS VIVERRINUS. 167

flow. This latter we regard as the morphological and physio-logical equivalent of the degenerated epithelial elements andblood extravasations met with in the pseudo-pregnant uterusof the marsupial.1 We are therefore in complete agreementwith those writers (Van Hervverden (12), Grosser (7),Hitschmann and Adler (15), Beard (3) and others) who holdthat " menstruation . . . is the expression of changes inthe uterine mucous membrane which are associated withpreparations for the reception of a fertilised ovum," andthat "menstruation itself is only a secondary process—a de-generation of the mucous membrane which from a failure ofpregnancy has not been able to fulfil its purpose " (Grosser,loc. cit. , pp. 97 and 102).

Our observations afford no support for the view that "men-struation is identical with ' h e a t ' " (Heape (11) p. 59), norfor the view " that menstruation in the Prireiates is thephysiological homologue of the pro-oestrum in the lowermammalia" (Marshall (18) p. 162).

As is generally recognised, menstruation is simply theoutcome • of degenerative uterine changes, and althoughthese are manifested at the end of the pro-oestral period inEutheria, our observations demonstrate that in the marsupial,they succeed both oestrus and ovulation, and cannot there-fore be considered as forming any part of the pro-cestralphase.

Now, bearing in mind the lowly position which theMarsupials occupy in the mammalian series, and taking into

1 Wiltshire states (27, p. 397), on the authority of Bartlett, that inkangaroos in the gardens of the Zoological Society, " a mattery, slimy "discharge from the cloaca, slightly tinged with a reddish colour, hadbeen observed by the keeper in females at the time of " heat." Wehave one record of the occurrence of an apparently corresponding dis-charge from the cloacal aperture in Dasyurus. The discharge isdescribed in our notes as " a whitish glairy secretion, consisting ofrefractive granules, round and angular," and is regarded as the secretionof the cloacal glands. We are not inclined to attach any importanceto its occurrence in the present connection, but it is quite possible thatthe cloacal glands may become more active during prooestrus.

168 J. P. HILL AND OHAS. H. o'DONOGHUE.

account the close similarity which is apparent between thecyclical changes in the pregnant and non-pregnant marsupial,it can hardly be doubted that the cestral cycle in this groupis not only simpler but much more primitive than that ofEutheria, so far as known. If this be admitted, then itfollows that, in the Eutheria, the inclusion in the pro-oestralperiod, of the degenerative uterine changes which conditionmenstruation is a purely secondary phenomenon, the resultof a thrusting forward of these events to a much earlierperiod in the cestrous cycle as compared with the naar-

So far, then, as concerns the uterine changes during pseudo-pregnancy, we conclude that these are represented in theEutheria by the alterations which'Occur in the mucosa duringthe latter part of pro-oestrus, and which, in some forms, cul-minate in menstruation.1

What induced this remarkable dislocation of events in theEutherian cestrous cycle is a problem by no means easy of solu-tion, but we venture to suggest that it may perhaps bebrought into relation with the omission or marked shorteningin the Eutheria, of the post-cestral period whereby ovulation isdirectly transferred to oestrus.

It is legitimate to assume that the shortening of the cyclein this way may have induced an increased growth of themucosa during the pro-oestral period, and that this growth inits turn may have directly conditioned the earlier occurrenceof the degenerative and regenerative changes, with the resultthat these latter now came into operation before instead ofaf ter the ovulation to which they were primitively related.Whether or not there be anything in these suggestions, therecan be little doubt that the precocious onset of the degene-

1 The liistological changes in the premenstrual uterus of the humansubject have been fully described by Hitschmann and Adler in theirimportant paper (15) published in 1908. These observers demonstratethat the cbanges in the premenstrual uterus are identical with thoseseen in the early pregnant uterus. To this paper, which contains avery full bibliography, the reader is referred for further details.

REPRODUCTIVE CYCLE IN DASYUBUS VIVERR1NUS. 169

rative changes in the mucosa, followed as they are by regene-rative growth, is of the nature of an adaptation, of obviousadvantage, as is generally recognised, from the point of viewof early placeutal formation (embedding, trophoblasticattachment, etc.).

Whilst, then, we believe that the uterine changes charac-teristic of the pseudo-pregnant period in the marsupial havebeen shifted forwards to pro-oestrus in the Eutheria, it remainsto be pointed out that amongst the Eutheria, alterations in theovary and mammary gland corresponding to those seen in thepseudo-pregnant marsupial have indeed been recognised, butdo not appear to have been brought into correlation' with thecestral cycle.

Various authors ( e .g . Ancel and Bourn (1) ) have describedthe formation of corpora lutea spuria, so-called, whichin animals ovulating spontaneously, last for a shorter timethan in the preguant female.

Further, as regards the mammary glands, it has been stated(Frank and Unger (6) and others) that a growth of theseoccurs during the oestral cycle, and according to Ancel andBouin (2), only after the formation of corpora lutea. One ofus (O'D.) has evidence confirmatory of this latter statement sofar as concerns the rabbit.

These changes in the ovary and mammary glands havethus retained their original position in the cycle, i . e . theyoccur after oestrus and ovulation as in the marsupial.

The Monoestrous and Polyoest rous Cond i t i ons .

According to Heape (11), monoestrous mammals are thosewhich experience a single oestrus during each sexual season,i. e. in which the anoestrons cycle alone occurs, whilst poly-cestrous mammals are those '' whose sexual season is occupiedby a series of dicestrous cycles, or in other words, those whoexperience a series of recurrent cestri" (p. 9). Heapequestions "if in the present state of our knowledge it ispossible to determine which is the original of these two con-

170 J. P. HILL AND OHAS. H. O'DONOGHUE.

ditions" (Heape, loc. cit., p. 18). We propose here toconsider this question very briefly in the light of such evidenceas we have of the oestral cycle in the Marsupials nnd Mono-tremes, which in respect of their reproductive phenomenagenerally (but making exception of the pouch of the Mar-supials) are undoubtedly more primitive than the Eutheria.

We have already shown earlier in this paper that the oestrouscycle in Dasyurus occurs only once in the breeding season,i. e. that Dasyurus is monaestrous. As regard's otherMarsupials we have no definite evidence, but our records, andthose of others (Semon (25) Caldwell (5)), indicate thatTrichosurus, Phascolarctus and Phascolomys breed butonce a year, each species having its own particular breedingtime, regularly recurring. We think in view of the posi-tive evidence we have concerning Dasyurus, the cautiousstatement of Heape (loc. cit . , p. 21) that " among certainwild animals which are known to undergo parturition onlyduring a very circumscribed time, the raonosstrous conditionmay be assumed as probable ," may be accepted

and applied to the forms mentioned.Selenka, however, makes the definite statement in regard

to D ide lphys mar sup ia l i s (23, p. 104)—"Die Brunst derWeibchen tritt normaler Weise nur ein Mai im Jahre ein.Wenn aber den Mutterthieren die Jungeu kurz nach demGebaren aus dem Beutel fortgenommen wurden oder wenn dieBegattung, was ofter vorkam, aus Mangel an Geschicklichkeitder Mannchen nicht gelang, so konnen die Weibchen 4-6Wochen spater zum zweiten male im Jahre briinstig werden,spatestens jedoch Anfang Juni."

further, with, reference toHyps ip rymn u s cun icu lus , hestates (24, p. 174)—"Die herrannahende Brunst . . . • zuverschiedeneu Jahreszeiten, im Fruhling, Herbst und Winterbeobachtet wurde."

Wiltshire, again (27, p. 397), states on the authority ofBartlett that the kangai-oos in the Gardens of the ZoologicalSociety "displa.y sexual excitement in September . . .and also in our spring month of April."

REPRODUCTIVE CYCLE IN DASYURUS VIVEEItLNUS. 171

We have had no experience of Didelphys, but cannot regardSelenka's statement as indubitable in the absence of anyrecord of the condition of the ovaries, whilst as regardsMacropus spp., we can only say that our records indicate thatin New South Wales at all events they certainly breed duringthe summer months (December to February). Irrespective,however, of the number of breeding seasons per year, wehave uo evidence as to whether the macropods are monoestrousor not.

As regards the Monotrenies, our own experience confirmsthe statement o£ Caldwell (5) and Semon (26)l to the effectthat the breeding season recurs bub once annually. In viewof the facts that at most two eggs in the single functionalovary reach maturity at the same time, and that the breediugseason is of such short duration that it would appear to beimpossible for a second set of eggs to become full-grownwithin its limits, we consider it justifiable to assume thatMonotremes are, like Dasyurus, monosstrous.

In view of these considerations and of others relative tothe breeding habits of reptiles which we do not think itnecessary to bring forward here, and in view, moreover, ofthe lowly position occupied by the monotrenies and marsupialsin the mammalian series, we find it difficult to avoid theconclusion that the monoestrous condition is the primitive one,and that the polyoestrous condition has been secondarilyderived from it. This latter condition where it occursamongst the Eutheria is obviously advantageous, since itpermits of the production of a larger number of young, or atleast provides greater opportunity for successful impregnation.

LIST OF REFERENCES.

1. Ancel.P., et Bouin, P.—" Sur les Homologies etla Signification desGlandes a Secretion interne de l'ovaire," ' Compt. rend. Soc. Biol..'fc. Ixvii, 1909.

1 Semon states (p. 8) that " die Monotremen haben in jedeni Jahrenur eine Brunst."

172 J. P. HELL AND CHAS. H. O'DONOGHUE.

2. Ancel, P., et Bouin, P.—" Le Developpement de la Glande mani-maire pendant la Gestation est determine par le Corpa Jaune,"' Compt. rend. Soc. Biol.,' t. lxvii, 1909.

3. Beard, J.—' The Span of G-estation and the Cause of Birth,'Jena, 1897.

4. Breselau, B.—"Die Entwickelung des Mammarapparates desMonstremen, Marsupialier und einiger Placentalier," 'Semon,Zool. Forschungsreisen im Australien, etc.," part iii, Bd. iv, 1912.

5. Caldwell, W. H.—" The Embryology of Monotremata and Marsu-pialia," part i, ' Phil. Trans. Roy. Soc.,' vol. clxxviii, B.

6. Frank, R. T., and Unger, A.—" An Experimental Study of theCauses which Produce the Growth of the Mammary Gland,"' Archiv. of Intern. Medicine,' vol. vii, 1911.

7. Grosser, O.—" The Development of the Egg Membranes and thePlacenta; Menstruation," 'Manual of Human Embryology,'Keibel and Mall, London, 1910.

8. Heape, W.—" The Menstruation of Semnop i thecus en te l lus , "' Phil. Trans. Roy. Soc, ' 1894.

9. "The Menstruation and Ovulation of Macacus rhesus,"' Phil. Trans. Roy. Soc, ' 1897.

10. " The Menstruation of Monkeys and the Human Female,"' Trans. Obstetrical Soc.'

11. " The Sexual Season of Mammals," ' Quart. Journ. Micr. Sci.,'vol. 44,1900.

12. van Herwerden, M.—"Beitrag zur Kenntnis des nienstruellenCyklus," ' Monatsschr. f. Geburtsh. u. Gyniikol,' Bd. xxiv, 1906.

13. Hill, J. P.—" On the Foetal Membranes, Placentation and Parturi-tion of the Native Cat (Dasyurus v iver r inus) ," 'Anat.Anzeig.' Bd. xvii, 1900.

14. " The Early Development of the Marsupialia, with SpecialReference to the Native Cat (Dasyurus v ive r r inus ) " ' Quart.Journ. Micr. Sci.,' vol. 56, 1910.

15. Hitschmann, F., and Adler, L.—"Der Bail der Uterussehleimhautdes gesehlectsreifen Weibes mit besonderer Berucksichtigungdes Menstruation," 'Monatsschr. f. Geburtsh. u. Gynakol.,' Bd.xxvii, 1908.

16. Marshall, F. H. A.—" The (Estrous Cycle and the Formation of theCorpus Lnteum in the Sheep," ' Phil. Trans. Roy. Soc.' B, vol.cxcvi, 1903.

REPRODUCTIVE CYCLE IN DASYURUS YIVERRINUS. 17 3

17. Marshall, F. H. A.—" The CEstrous Cycle in the Common Ferret,"'Quart. Journ. Micr. Sci.,' vol. 48, 1904

18. ' The Physiology of Reproduction,' London, 1910.19. and Jolly.—" The CEstrous Cycle in the Dog," ' Phil. Trans.

Roy. Soc.,' B, vol. cxcviii, 1905.20. O'Donoghue, C. H.—" The Growth-changes in the Mammary

Apparatus of Dasyurus and the Relation of the Corpora Luteathereto," ' Quart. Journ. Micr. Sci.,' vol. 57,1911.

21. " The Corpus Luteum in the Non-pregnant Dasyurus andPolyovular Follicles in Dasyurus," ' Anat. Anzeig.' Bd. xli, 1912.

22. Sandes, F. P.—"The Corpus Luteum of D a s y u r u s v ive r r inus ,with Observations on the Growth and Atrophy of the GraafianFollicle," ' Proc. Linn. Soc, New South Wales,' 1903.

23. Selenka, B.—' Studien iiber Entwicklungsgeschichte der Thiere,'Bd. iv (Heft 1 and 2) "Das Opossum (Didelphys vir-g in iana ) , " Wiesbaden, 1887.

24. Ibid., Bd. v (Heft 1), " Beutelf uchs und Kanguruhratte, etc,"Wiesbaden, 1892.

26. Semon, R.—' In the Australian Bush,' Macmillan & Co., London,1899.

26. " Beobachtungen iiber die Leberisweise und Fortpflanzungder Monotremen, u.s.w.," ' Zool. Forschungsreisen im Australien,'Bd. ii, Lief 1,1894.

27. Wiltshire, A.—" Lectures on the Comparative Physiology of Men-struation," ' British Medical Journal,' March, 1883.

28. Thomas, Oldfield.—"Cat. of the Marsupialia and Monotremata,"British Museum (Nat. History), 1888.

174 J. P. HILL AND CHAS. H. o'DONOGHUE.

EXPLANATION OF PLATES 6, 7 AND 8.

Illustrating Prof. J. P. Hill and Dr. Chas. H. O'Donoghue'spaper on "The Reproductive Cycle in the Marsupial,Dasynrus viverrinus."

[All the photo-micrographs are from utei'iof Dasynrus and-are takenat a magnification of 30 diameters except figs. 9 and 10, the magnifica-tion of which is 250 diameters.]

PLATE 6.

Fig. 1.—Uterus, pro-oestral Case 1 (17 . vi. '99).Fig. 2.—Uterus, pro-cestral Case 2 (2 . 21 . v. '03).Fig. 4.—Uterus, post-oestral Case 1 (8, 23 . vii. '02).Fig. 5.—Uterus, early pregnant Case 1 (15, 19.vii. '01).

PLATE 7.

Fig. 3.—Uterus, cestral Case 1 (12 . vi. '02).Fig. 6.—Uterus, pseudo-pregnant Case 2 (vii. '06).Fig. 8.—Uterus, pseudo-pregnant Case 4 (31. vii. '99).Fig. 9.—Portion of mucosa of same uterus as fig. 6, showing uterine

glands lined by low cubical epithelium and with their lumina filled bymasses of cells.

PLATE 8.Fig. 7.—Uterus of pseudo-pregnant Case 3 (1900).Fig. 10.—Portion of a uterine gland from the same uterus as fig. 3,

showing the ciliated gland epithelium.

. VoL. 59J.&&L. 6.

5.

C.H. O'D. plioto.

C.H. O'D. photo