97° Journal of Agricultural Research voi. v, No. 21 
form, many of the aphides leaving the plants and dying before the repro¬ 
ductive period was finished. Consequently 42 young is probably below 
the average under natural conditions. 
The young are produced in groups of varying numbers and with unequal 
periods between the groups. In a general way an adult will produce a 
group one day and rest the next, but often the rest period will be 
longer and sometimes shorter. Individual mothers vary greatly in their 
rate of reproduction from the average rate. Some stem mothers ceased 
to reproduce for 2 or 3 days between some groups, while others never 
rested long. The greatest number of young produced in 24 hours was 
9, one insect producing this number at two different times. In 4 days 
22 were produced by one mother. The average daily production was 4.2. 
LONGEVITY 
The greatest length of life observed was 20 days. This is undoubtedly 
much below the true maximum and probably somewhat below the average. 
In the case recorded the insect produced young up to the last day. 
The first stem mothers were observed on April 8 and the last May 6. 
Under natural conditions this period may perhaps be a little longer. 
SUMMER FORMS 
NUMBER OF FORMS 
Beginning with the second generation and continuing until the sexes 
were produced, the writers found three adult forms to be present. The 
most abundant form was the wingless viviparous female. This occurred 
in every generation, and, with the exception of the second, always out¬ 
numbered the other forms present. It would often appear, in definite 
lines of descent, for several generations without being accompanied by 
winged insects. In fact, one purely wingless line was carried from the 
stem mother to the sexes, although in this case winged forms sometimes 
occurred as sisters or cousins. 
On the other hand, the winged form was much more abundant than 
seems to be the case in most of the other species which have been studied. 
Winged insects were obtained in every generation from the second to the 
sixteenth, inclusive, although they became rare after the thirteenth 
generation. 
The third form, the intermediate, occurred in 16 experiments, the first 
occurrence being in the third generation and the last in the twelfth. 
In all, there were from 7 to 17 generations of the summer forms, the 
number depending upon whether the first or the last young were taken as 
mothers in each generation. In view of the fact that we have found 
winged forms to occur so commonly, it is difficult to understand how 
Smith (9) could have come to the conclusion that no winged insects 
