272 
Psyche 
[June-September 
two pairs of abdominal spiracles show signs of dorsal migration 
(Henry, 1976, Fig. 10). Finally, the Oligocene Neadelphus protae 
MacLeod displays ventral spiracle location and is devoid of ventral 
scoli, vestigial or otherwise, but those it possesses number twelve 
rather than the ten of Ululodes and its relatives; it also shows no 
setal modification (MacLeod, 1970; Henry, 1976, Fig. 9). 
MacLeod (1970) interprets the secondary abdominal scoli of New 
World neuroptyngines as having migrated from the ventral series of 
a non-flattened nyinphid-like ascalaphid ancestor. My own view, 
based on analysis of changes in spiracle location and comparison of 
scolus shapes, is that the primary rather than secondary scoli of 
these neuroptyngines have been derived from the ventral series, and 
that the primary abdominal scoli of Neuroptynginae and Ascala- 
phinae are therefore not homologous (Henry, 1976). 
The Mkuze specimen described here shares its general head 
capsule morphology and its twelve pairs of primary body scoli with 
Neadelphus and known extant Neuroptynginae and Old World 
Ascalaphinae; in both respects, it is a primitive or generalized 
ascalaphid. Its most remarkable feature, however, is the double row 
of scoli on each side of its abdomen with spiracles placed laterally 
between the rows: although Ascalaphus shows traces of the ventral 
series and New World neuroptyngines bear the latter in the same 
plane as the dorsal series, no larva possesses such a fully developed 
double series nor so closely approaches MacLeod’s hypothetical 
nymphid-like ancestor as this specimen. 
The setal patterns on the scoli of the specimen may help to 
homologize these protuberances in this and other ascalaphid larvae. 
For example, it is not known which (if either) of the two pairs of 
scoli on the mesothorax or on the metathorax represent the dorsal 
series, since spiracles are not present on either segment and all 
known taxa showing the condition bear both pairs in the same 
plane. The Mkuze larva possesses sharp-tipped setae on the tips of 
the first and third pairs of thoracic scoli and on all ventral scoli, 
suggesting a ventral origin for the more anteriorly placed pair of 
scoli on each thoracic segment. Actually, the same relationship is 
also present in Ascaloptynx: the first and third thoracic scoli 
resemble in shape those deduced to belong to the ventral abdominal 
series (Henry, 1976, Fig. 5). It then follows that, in neuroptyngine 
larvae bearing all acoli in a common plane, the anterior scoli on 
each thoracic segment did not originate from the same series as did 
