counter balance to mortality in preserving a pop- 
ulation. Davis (28) has treated thoroughly this 
concern with reproduction as it pertains to Norway 
and roof rats. 
My studies of the Towson population of Norway 
rats made me realize that phenomena relating to 
reproduction can be given an entirely different 
focus. This focus concerns the conditions sur- 
rounding the maturation of the individual as these 
conditions modify physiology and behavior, which 
in turn alter reproductive performance. In other 
words, reproductive performance becomes a cri- 
terion for judging the approach to the optimum 
by environmental and social conditions. 
High reproductive performance should indicate 
favorable conditions in so long as the species does 
not exercise conscious control of the frequency of 
conception. If we know the environmental, social, 
and cultural conditions which foster high repro- 
ductive states in lower mammals such as the rat, 
we can examine them for general principles appli- 
cable to producing less stressful conditions of living 
by man, irrespective of the associated human 
reproductive performance. In other words, the 
general kinds of conditions which assure a high 
reproductive rate for rats should be indicative of 
the type of environment providing the minimum 
of stress for humans. 
The Towson data (pp. 152 to 160; 203 to 216) 
provide an insight into some of the conditions 
associated with reproductive success. Such success 
involves appropriate interaction with males, culmi- 
nating in insemination; fertilization of ova, non- 
resorbtion of embryos; and appropriate maternal 
behavior terminating in weaning of the young. 
However, the Towson data (pp. 158 to 160) are 
too few to provide complete insight into the matura- 
tion of reproductive performance. 
Several sources of such quantitative data are 
available in the literature. These will be examined 
for the light they throw on maturation of repro- 
ductive performances and its modification by 
environmental conditions. 
In the study of wild populations, it is customary 
to trap large samples throughout the year. In 
lieu of exact knowledge of age, characteristics of 
the sample are usually listed according to length 
or weight classes. However, insight into the 
maturation of reproductive processes demands a 
knowledge of approximate age. It is obvious from 
figures 1 30 to 1 33 that equal intervals of length or 
weight do not correspond to equal intervals of age. 
Therefore, in order to assign approximate ages to 
the midpoints of class intervals of length in milli- 
meters for data in the literature, the 426 measure- 
ments of Towson females falling between 170 and 
and 270 mm. were reexamined. For each 10 mm. 
interval of length I calculated the mean and 
standard deviation of the age in days. Approxi- 
mate ages and their standard deviations as read 
from a smoothed curve drawn through these points 
are shown in columns 2 and 3 of table 68. 
One-hundred and four measurements of nose-to- 
anus length were available for the Towson colony 
rats for which there was another measurement 
within seven days or less. Differences between 
these two measures represent variability in the 
measurement rather than changes due to growth. 
The midpoint difference, or error, was 4 mm. 
Thus 0.4 of the readings in any 10 mm. class interval 
should actually have been in the next smaller or 
larger class interval. For this reason the actual 
range in age for each class interval of lengths must 
be more nearly approximated by rather than 
± 2 " 
Two kinds of data on reproductive performance 
are available from samples of populations of wild 
rats: (a) the proportion which are parous, that is 
they are able to bear litters (i.e. as judged by the 
fact that they are visibly pregnant or have placental 
scars from prior litters), and (b) the proportion 
which are visibly pregnant with the naked eye 
upon external examination of the exposed *uterus. 
Davis and Emlen (66) present data on parity of 
out-of-doors, urban rats in Baltimore. Estimates 
of the proportion parous are given in column 4 of 
table 68. These estimates are based upon a 
smoothed curve drawn through their data on 430 
females. Some females first conceive while quite 
young, whereas for others this event is delayed for 
many months. At 165 days of age 0.5 of the rats 
have still not conceived, despite the fact that all 
have probably been ovulating for a long time. 
Davis (28) shows that 0.5 of the rats are ovulating 
at a length comparable to 40 days of age. There- 
fore, for the average female conception is delayed 
for 125 days beyond first ovulation or 80 days 
beyond the youngest rats to conceive. Conception 
seemed to be inversely proportional to stressful 
social conditions in the Towson population. There- 
fore, I suspect that variability in the social experi- 
ence and effectiveness of other behavioral 
adjustments to environmental conditions operated 
in the urban populations of Baltimore rats to pro- 
676-708 
265 
