126 Psyche [August 
embryo grows by the absorption of food and fluids from the sur- 
rounding body fluids of the host. 
The first stage larva proved to be of unusual interest, and of 
specialized structure as is so frequently the case in first stage larve 
of hymenopterous parasites. A short description of it with an out- 
line drawing will follow later in this paper. Its extraordinary tail- 
like appendages are probably respiratory in function, and the 
comb-like arrangement of bristles on the dorsum seems to indicate 
that it may thereby be able to move about among the organs and tis- 
sues of the host. The assumption that the bristles may aid in its 
locomotion is strengthened by the fact that they all point backward, 
are undoubtedly of sufficient size and firmness, and occur most 
abundantly on the posterior part of the body where they would be 
most useful. 
After the first molt the larva loses its extraordinary appearance 
and assumes the form of the ordinary hymenopterous larva. During 
the second stage it increases rapidly in size and begins to make in- 
roads upon the tissues of the host, though as yet carefully avoiding 
the vital organs. Finally in the third and last stage the larva quite 
destroys all the organs of the host, including the numerous embryo 
aphids which the mother aphid carries in her abdomen, and after 
sucking up the last juices, it leaves its host as a perfectly dry shell, 
in order to spin underneath a characteristic white cocoon, which 
has been figured and described by previous authors.* 
Specimens of the Praon under observation were sent to the Bureau 
of Entomology at Washington and identified by Mr. H. L. Viereck 
as Praon simulans Prov. The aphid acting as host was likewise 
determined by Mr. Theo. Pergande to be the common rose-aphid 
Macrosiphum rose L. Another aphid common in the local green- 
house on carnations also proved to be host of this species though less 
frequently attacked. 
From cocoons of Praon were obtained several species of secon- 
daries. One a species of Encyrtid was obtained under such condi- 
tions as to make it extremely probable that this species does not 
attack the cocoons of Praon directly, but lays its eggs in the body of 
the aphid, where after hatching the larva would seek out the Praon 

1 For figure of cocoon see Howard, Insect Life, Vol. 4, p. 196. 
