6 BULLETIN 1487, U. S. DEPARTMENT OF AGRICULTURE 
This general external feeding by the hyperparasitic larvae implies 
the deposition of eggs directly upon the primary larvae inside. their 
cocoons or puparia. Although this is the usual manner of ovi- 
position, there are numerous interesting exceptions. Those species 
of the genus (Chalcis) Brachymeria which are parasitic upon Tach-- 
inidae deposit their eggs inside the tachinid maggots before the 
latter have issued from their hosts, which are various lepidopterous 
larvae and pupae. Without this adaptation in oviposition the 
Brachymeria would succeed in finding but few of their tachinid 
hosts, since these usually enter the soil immediately upon emerging, 
and so are rather well protected throughout the puparial period. 
No such advantage, however, is derived by species of the ichneu- 
monid genus Mesochorus from a similar habit of oviposition; for 
these attack hymenopterous species that form their cocoons in the 
open where they are easily accessible. Any benefit that species of 
Mesochorus enjoy from this manner of attack must consist in the 
advantage that early possession of the primary parasite gives them 
in the competition with other hyperparasitic forms for those partic- 
ular hosts. Certain Cynipoidea, represented by species of Charips 
that are hyperparasitic upon aphids, and a species of Tetrastichus 
which is abundantly reared from cocoons of Apanteles glomeratus 
L., also attack the host of the primary in this manner, for the sake 
of the parasite it may already contain. A better acquaintance with 
the biology of hyperparasitic species will doubtless reveal a wider 
occurence of this habit. None of the secondaries just mentioned 
can develop as primary parasites upon the hosts attacked; the few 
observations that have been made indicate that they do not even 
deposit eggs unless a primary parasite is present within, and then 
always place their eggs inside the body of the primary. 
One of the most interesting types of indirect parasitism among 
hyperparasites is that exhibited by Perilampus as described by Smith 
(34). The newly hatched planidiumlike larva of Perilampus must 
attach itself to a passing host of one of the parasites upon which it 
can develop, and must then bore inside in the hope of finding the 
desired parasite. It does not find its food supply ready at hand, on 
hatching from the egg, as practically all other parasitic species do. 
Since only a very small part of the Perilampus planidia succeed in 
locating the larvae of primary parasites in this roundabout manner, 
an enormous capacity for egg development and deposition by the 
parent females is required. A habit very similar to that of Perilampus 
has been observed by Clausen (2) in Schizaspidia, a member of the 
closely related family Eucharidae; but in this case the parasite is 
primary. 
FEEDING OF ADULT HYPERPARASITES 
The habit with certain parasitic insects, particularly chalcidoids, 
of feeding at the puncture holes made by the ovipositor has been 
discussed by various writers, some of the more interesting observa- 
tions being those recorded by Marchal (24, 25), Howard (17), Doten 
(S), Johnston (20, 21), and Rockwood (33). In all these cases primary 
parasites were concerned. But hyperparasites have the same habit, 
which, as Howard suggested, probably ‘‘will be found to be quite 
widespread.”’ Nearly all the parasites of Apanteles melanoscelus 
treated in this bulletin have been observed feeding at the punctures 
