101 
The five known forms of this insect are as follows: the egg, 
wingless pseudogyne or viviparous female, pupa, winged pseu- 
dogvne, and the wingless sexed form, consisting of true oviparous 
females and of males. The eggs have been found in spring, from 
which the young root lice have hatched in the first part of 
April. These are viviparous, and give birth to another genera¬ 
tion of the same kind, and successive generations follow during 
the entire summer. Young have been taken in nearly every 
collection here during the summer months. 
In the fall, nsualiy in October and November, the viviparous 
form gives birth to true sexed individuals, males and oviparous 
females. These copulate and the females lay their eggs in the 
loose earth around the roots of plants in the ant burrows in 
which they are kept. The ants keep these eggs during the win¬ 
ter and spring, and from them in due time the young viviparous 
root lice hatch. Thus the circle of life continues. It is shown, 
however, by our collections that some of the lice also live dur¬ 
ing the winter on the roots of plants, in the formicaries of ants, 
most of these being viviparous and young; but the oviparous 
form sometimes occurs there during the winter or very early 
part of the spring. 
description.* 
This plant louse is closely allied to certain species commonly 
placed in Tjmhea, a genus of the Rhizobiinse, which is a sub¬ 
family whose members have not hitherto been known to acquire 
wings. A careful study of the winged individual bred from a 
pupa shows that it is a new generic type most nearly related 
to Hormaphis, of the Pemphiginae; and as Tychea seems to 
contain at least two dissimilar groups of species, whose only 
bond of union is a resemblance in the number and length of the 
antennal joints, it will be well to divide the genus, retaining the 
old name for the typical species, such as phaseoli and brevicor- 
nis, and placing seiarise and its allies together with the present 
species at the foot of the Pemphiginse, as a new genus, for 
which the name Geoica is now proposed. Here also belongs 
the species described by Buckton as Endeis carnosa, which surely 
is not an Endeis at all. Tychea panici, Thos., is apparently a 
young Rhizobius. 
geoicaI n. gen. 
Antennae 5-jointed, not annulated, first and second joints short, 
third longest, fourth and fifth shorter, subequal, often connate, 
the fifth with a short thick spur at tip. Sensoria present on the 
third antennal joint of the winged individual, and in all of the 
forms at the apex of the fourth joint and base of spur on fifth, 
the latter sensorium lunate in the wingless individuals. None 
* By C. A.. Hart. 
+ Ge, earth; oikos, house. 
