ON UROTHRIPS PARADOXUS. 
13t 
portance in classification. This is especially so in the snb-order Tere¬ 
brantia, the number and arrangement of the bristles on the pro- 
thorax, or the absence of them, being used to some extent as generic 
characters. 
In the suborder Tubuli fera five important pairs of prothoracic 
bristles are recognized, whilst each segment of the abdomen bears a 
few setæ, which are small on the anterior segments, and increase in 
size and prominence towards the tube, the ninth segment always 
bearing a series of hairs around the hind-margin, which are often 
longer then the tube. At the tip of the tube there is invariably a 
circlet of long, slender hairs, which appear, however, to be of a diffe¬ 
rent character to the above-mentioned setæ, and will be more parti¬ 
cularly dealt with herein under the heading of «the eleventh abdominal 
segment». 
The characters of the prothoracic and abdominal chætotaxy will 
ultimately prove, I believe, to be of undoubted taxonomical value in 
the Tubuli fera as well as in the Terebrantia . 
Duly considering the above facts, Urothrips presents a most 
interesting condition in being entirely without the above-mentioned 
setæ. Under high magnification, and in certain lights, however, seta*pits 
can be detected in the chitinous skeleton of the head, thorax and 
abdomen, though for the greater part these are irregularly placed, 
minute and of no .importance. Under still higher magnification a minute 
black spot may be seen in the centre of more than one of these minute 
seta-pits, whilst what appears to be an extremely short and most 
minute hair, is seen to spring from one or two of the other pits. 
Whether these are the remnants of hairs, or poorly developed ones or 
wether the appearance is altogether deceptive, and only an optical illusion 
caused by certain lights, it is difficult to say. Knowing that the Tubu- 
liferous Thysanoptera, without exception, possess more or less promi¬ 
nent spines on the hind-margin of the ninth abdominal segment, one 
would naturally expect to find corresponding seta-pits in Urothrips. 
But this is not the case, the only pits of any value being found in the 
prothorax, where the mid-lateral and posterior-marginal pairs are pre¬ 
sent, though not large, nor easily detected (PL III. fig. 1). 
Maxillary and labial palpi. 
The form and segmentation of the maxillary and labial palpi of 
the known Thysanoptera form useful characters in the division of the 
families. The labial palpus is always two-segmented in the Tubulifera 
9* 
