4 BULLETIN 1487, U. S. DEPARTMENT OF AGRICULTURE 
Davis (7), who discusses the rhipiphorid Macrosiagon pectinatus 
Fab. as a parasite of Tiphia; and Fiske (10), who records the cucujid 
Catogenus rufus Fab. as a parasite of a braconid primary parasite. 
Of course it must be understood that it is impossible to designate 
absolutely a certain group of species as hyperparasites as distinguished 
from primary parasites. As will be pointed out in greater detail 
farther on, there are many species that sometimes act as primary 
parasites whereas under other conditions they are hyperparasitic. 
SPECIFIC HOST RELATIONS 
In general hyperparasites are undoubtedly less discriminatory than 
primary parasites as regards the selection of hosts. Certain forms, 
such as species of the ichneumonid genera Hemiteles and (Pezo- 
machus) Gelis and the chalcid Dibrachys boucheanus (Ratzeburg), 
attack practically anything resembling the cocoons of hymenopterous 
primary parasites; and some species parasitize dipterous and hymen- 
opterous parasites alike. Species of the chalcid genus Melittobia 
apparently attack a great variety of Hymenoptera, both parasitic 
and free-living (13, 14), and are also sometimes serious enemies of 
Tachinidae (19). The European eupelmid Hupelminus saltator Lind., 
which is now established in the United States, develops upon such 
dissimilar hosts 'as the Hessian fly and species of Apanteles. And 
Smith (34) records species among the Tachinidae, Ichneumonidae, 
Braconidae, and Chalcidoidea as hosts of Perilampus hyalinus Say. 
Many other illustrations of such indiscriminate parasitism could be 
given. Though a distinct preference is usually exhibited for one 
general type of host, this ordinarily includes a very wide range of 
species. Only rarely does a hyperparasite confine itself rather closely 
to certain few host species, as in the case of (Chalcis) Brachymeria 
compsilurae Crawf. and Monodontomerus aereus Walk., which are essen- 
tially parasites of a few particular Tachinidae having similar habits. 
As arule hyperparasites are better able than primary parasites to adapt . 
themselves to a great variety of hosts when the preferred species 
are not available in sufficient numbers. This doubtless accounts for 
the continued abundance of certain hyperparasites in a given locality 
irrespective of the presence of the primaries that are their preferred 
hosts. It will also account, as Howard (1/8) has indicated, for the 
heavy parasitism upon primary parasites introduced from another 
country, even though these are imported unaccompanied by any of 
the hyperparasites that attack them in their native habitat. 
The term ‘“‘secondary parasites’? is sometimes applied to hyper- 
parasites aS a whole, and after making certain allowances may 
be considered strictly correct. As suggested by Fiske (1/7), true 
tertiary or quaternary parasitism, or parasitism of even higher 
degree, is of rare occurrence. It is doubtful if any species are obliga- 
tory tertiary parasites. Some, like the two species of the eulophid 
genus Pleurotropis, which are discussed subsequently in this bul- 
letin, are evidently preferably tertiary; but they can, and some- 
_ times do, act as true secondary parasites. More commonly species 
that are normally primary develop as secondaries; and conversely 
some of those that are usually secondary become primary under the 
proper conditions. As an illustration of this latter adaptability the 
behavior of EHupteromalus nidulans (Foerst.) is interesting. This 
