Dec. ii, 1916 
Macrosiphum granarium 
475 
sexes. The writer has taken both pink and green individuals in the 
summer and kept the progeny of each isolated in rearing cages until 
fall. The sexes appeared among the descendants from the pink indi¬ 
viduals, but very sparingly from the descendants of the green ones. 
This may not hold in every instance, but it has been the experience of 
the writer that the sexes can be obtained with certainty by starting 
with the pink summer forms. 
If the cages are examined closely when the sexes begin to appear in the 
fall, two distinct types of adults will be noted. One is the usual green 
form (PI. B, 2) and the other will have a pinkish tinge (PL B, 1). If 
the pink wingless individuals are isolated it will be found that they 
produce two kinds of young, one slightly tinged with pink (PL B, 4) 
and another a.deep pink (PL B, 3). The color of the mother after 
she begins producing young is due in the main to the pinkish young 
showing through the body wall. The slightly pink individuals are 
usually produced first. The offspring of the pink wingless individuals 
usually all become winged, the deep-pink individuals developing to 
winged males, while the pale-pink winged individuals are viviparous 
and produce the wingless, yellow, oviparous females (Pl. B, 5). This 
fact was not known definitely by the writer until the fall of 1909, when he 
isolated a few pink wingless viviparous females in order to learn what sex 
their offspring would be. In every case the results were as just stated. 
In the fall of 1910 pink wingless individuals were again isolated for 
observation on their progeny. The results obtained entirely corrobo¬ 
rated the data of 1909. In the fall of 1911 a large series was isolated 
as in 1909 and 1910. A heavy storm accompanied by very low tempera¬ 
tures put an end to the observations before the data were complete, 
killing all individuals under observation, since the rearing cages offered 
little natural protection from cold. 
Since 1909 and 1910 the writer has found that the winged viviparous 
females of this series 1 may produce only viviparous individuals in some 
cases or both oviparous and viviparous, or may produce only the ovip¬ 
arous females. The males are produced only by the wingless pink 
viviparous females and the oviparous females are produced only by the 
winged adults that develop from the slightly pinkish young. In other 
words, the males are sons of the pink viviparous females and the ovip¬ 
arous females are the granddaughters. In no case are oviparous 
females and males produced by the same mother. The oviparous 
females might be termed nieces of the males; they are never sisters of 
the males. 
The following outline will illustrate more clearly the sequence of the 
sexes: 
1 The offspring of the pinkish wingless viviparous females of the autumn forms. 
4 
