to nine weeks after the beginning of the breeding season. Reynolds assumed that these were individuals that had failed to find mates during the first oestrus of the season and that after completing the regular dioestrus of about 28 days they had then mated and borne young. Wiseman and Hendrickson (1950:333) in southeastern Iowa recorded a female with a litter no more than two days old on February 23, and several other females with young were estimated to have borne litters at approximately this same date, while still others bore litters as late as early March. Two lots of small young found in early June may have been second litters.

For the region represented by the present study, the data indicate a breeding season with later onset and sharply circumscribed limits as compared with an earlier onset and less circumscribed limits in Texas, central Missouri, and even southeastern Iowa, which is a little farther north. The available data indicate that there are two distinct and well-defined breeding seasons in the course of the annual cycle on the University of Kansas Natural History area. The whole population, including young of the preceding year, some still far below average adult size, breeds from about the middle of February into early March, and first litters are born mainly in early March. Individual females may vary as much as two to three weeks in the time of breeding, and varying weather conditions from year to year may hasten or delay onset of the breeding season. Data are recorded below for all females caught in March that were carrying litters.

None of the females trapped in February was carrying young in the pouch, but probably some early litters are born in the last week of February or even earlier. By late March most of the females are carrying young in their pouches, and those which do not have young, have their pouches enlarged and vascularized for accommodation of the young. Presumably such females have already borne young and then lost them. Nearly all the litters seen in the latter half of March had young that were much larger than at birth.

Of 13 females examined in April, 12 were carrying young, and the remaining one was known to have been carrying a single young on March 1, but had lost it. Eleven females were examined in May, four of which were the same ones examined in April. Eight of the eleven females were carrying young; of the remaining three, one had lost the litter of young that it had been carrying when trapped in April. Two had empty pouches on May 19 and 20, but probably had successfully reared the litters of young which they had been carrying when trapped in April. The young of all those females trapped on different dates in April and May were in stages of growth indicative of birth about the first week in March. The latest date on which a female was recorded with first-litter young in the pouch was May 22, 1951, and these were the largest pouch young observed. Their eyes were recently opened, they were estimated to weigh 60 grams each with hind feet 20 mm. long. Young continue to grow rapidly after leaving the female's pouch. A young female caught on June 16, 1949, weighed 126 grams. For seven young caught on July 5 and 6, 1952, weights and hind-foot measurements were, for males: 660 grams, 52 mm.; 560 grams, 46 mm.; 550 grams, 48 mm.; 450 grams, 44 mm.; 370 grams, 44 mm.; 330 grams, 37 mm.; and for the one female: 430 grams, 46 mm.

The wide variation in size in this small group of young of nearly the same age is noteworthy. Size and condition of the females carrying them, number of competing litter mates, and early success or handicap in independent life causes so much divergence in size that at the age of four months some young are twice as large as others.

By late fall the young grow to small-adult size. For example, the female that weighed 126 grams when first caught on June 16, 1949, was recaptured on November 29, 1949, and on that date weighed 1710 grams.

A second breeding season ensues soon after the young of the first litter leave the pouch, and these young probably soon learn to shift for themselves. Second litters are usually born in early June. On June 14, 1952, a female was taken with young only a few days old in her pouch. On July 5, 1952, two females last taken on May 19 and May 20, with their pouches recently vacated by first litters, were found to have young the size of half-grown mice, evidently two to three weeks old. In the months of October, November, December and January, a total of 11 young, thought to represent second litters, were taken. Dates of capture, weights in grams and sexes were as follows:

The hind foot measured 48 mm. and 51 mm., respectively, in the young weighing 630 grams and 660 grams. These young, born in early summer have grown, by October, to a size comparable with that attained in July by young of the early spring litters. The variation in size is also similar but with a little wider range. The summer breeding season may be somewhat more protracted than the breeding season in early spring.

Too few females were caught in summer to compare the summer breeding season with the early spring breeding season, with respect to size of litters, percentage of non-breeders, and other factors which might affect the size of the crop of young produced. It is not clear why, among opossums trapped in winter, the young born in early spring outnumber those born in early summer by about four to one. Some females are eliminated after rearing the first litter, and others, exhausted by rearing large first litters may fail to participate in the second breeding season. However, it seems that the young of the summer litters must be subject to other unusual and selective mortality factors which eliminate most of them by fall. That such factors vary from year to year is indicated by the changing ratio of summer-born young to other opossums in each of the three winter seasons when trapping was carried on.

Hartman (1952) has summarized his own findings and those of other authors regarding the embryology, birth, and early development of the opossum, and has corrected numerous popular misconceptions. He states that an average litter consists of about 21 eggs, but mentions much larger litters of up to as many as 56. However, many of these may fail to develop. The female normally has 13 functional nipples in her pouch and each one accommodates a single young. Excess young beyond this number are doomed, and soon perish from starvation if they reach the pouch after all the nipples are occupied. None of the females examined in the present study had a full complement of 13 young. Under unfavorable conditions, most or all of the young may fail to make the trip from the vaginal orifice to the pouch. Also, the pouch young are subject to heavy mortality, but observations concerning the time and cause of mortality are lacking.

Lay (loc. cit.) found an average of 6.8 pouch young in 65 litters examined in eastern Texas; Reynolds found an average of 8.9 (5 to 13) in 42 litters from Boone County, central Missouri; Wiseman and Hendrickson found an average of 9 (6 to 12) in southeastern Iowa. In the present study, 28 of the female opossums examined were carrying litters in their pouches, and all these females were caught in the months of March, April, May, June and July. The number of young varied from one to 12. Seven females each had seven young, six each had eight, three had six, three had five, and there were two each with nine, 10, and 12 young, and one each with one, four and 11 young. The average was 7.4 per litter. On several occasions females captured with young in their pouches and recaptured one or more times within a few weeks, were found to have lost some or all of the young. Some of the females examined probably had already lost parts of their litters. For instance, the female recorded with just one small young on March 1, probably had lost most of her litter and when recaptured a month later she did not have any young.

Nineteen yearling opossums were taken in the fall-winter-spring season of 1951-52; 42 per cent of the total, and 67 per cent of the females were individuals marked as pouch young the preceding spring. In the course of live-trapping, that spring, some first litters may have been missed. No second litters were marked because trapping was not continued into June and July when second litters are being carried by females. These figures suggest that the breeding population of females on an area consists chiefly of those born there the preceding spring.

Sex ratio of opossums trapped was approximately 1:1; 59 males to 58 females. Age groups for opossums caught in the three seasons are shown in the following tabular fashion. For a few individuals age status was doubtful.

In the 1950-51 season, small young of the summer brood seemed unusually numerous. In the 1951-52 period, young of both age classes were relatively scarce and old adults made up an unusually high proportion of the population. Excluding the 14 marked pouch young that were later recaptured, there were only four of the total of 106 that were trapped in each of two seasons. One young less than a quarter grown, that was accidentally caught in a live-trap set for woodrats, was recaptured as a breeding adult the following winter. An adult male and two adult females each caught in the 1949-50 season were each recaptured repeatedly in the 1950-51 season. Ninety-five per cent replacement of the breeding population by the following breeding season is indicated by our figures. Only 3 (or 5 per cent) of the individuals of the population trapped and marked in the season of 1949-50, were recaptured among the 62 opossums recorded in the two subsequent seasons. Various mortality factors including predation, disease, and accidents account for some