1927] 
On the Affinities of the Gryllohlattidce 
37 
the hind pair modified for leaping: tarsi never more than 4- 
jointed. Sound producing organs almost always present. Cerci 
never multiarticulate. Ovipositor well developed and almost 
always exserted. Penis bilobed and symmetrical. 
On the characters enumerated above it may be said that the 
Grylloblattidae only differ from the Cursoria in the possession of 
a well developed exserted ovipositor. It is, however, necessary to 
examine certain other criteria and not base a definite conclusion 
solely upon this diagnosis. Furthermore, several of the features 
stressed by Crampton in maintaining his point of view need 
some comment. 
(1) Absence of ocelli. This character is of little phylo- 
genetic value as ocelli tend to degenerate or disappear in various 
Orthoptera. Thus, among the Cursoria they are reduced or 
wanting in some apterous Blattidae as well as being absent in 
many Phasmidse. Their absence, therefore, is no criterian of 
affinity with the Saltatoria. 
2; Multiarticulate cerci only occur in the Blattidae and 
the related family Mantidae. Unjointed setiform cerci are found 
in the Gryllidae but there is no indication that this superficial 
similarity to the same, but multiarticulate, organs in the Gryllo- 
blattidae is anything more than a parallelism, induced perhaps, by 
a similar terrestrial life. 
(3) Dr. Crampton has already pointed out the structural 
similarity that exists between the terminal abdominal segments 
of the Grylloblattidae and Mantidae. He has likewise stressed 
the similarity of the coxae, trochanters and divided trochantins, 
as seen in the Grylloblattidae and Blattidae, so no further com- 
ment here is needed. 
(4) The various Isopteran characters displayed in the 
Grylloblattidae, and emphasised by Crampton, afford indirect 
support to the relationship of that family with the Blattidae. 
It is generally accepted by such competent authorities as Holm- 
gren and others that the Isoptera are closest related to the 
Blattidae on the sum total of their characters. Evidences of 
affinity of the Grylloblattidae with the Dermaptera and Embiop- 
tera have very little bearing upon their systematic position within 
the Orthoptera. 
