I'ii pal suspension of Thais. 205 



position that may be described as in front of the larva, 

 the head being thrown back, so that the legs are used as 

 hands, one might say, to hold it up. Not, however, the 

 claws, but the thick bases of the legs are used, the silk not 

 being on the legs proper, but rather in the incision in front 

 of them. This is the position when the spinneret is at the 

 middle of the girth. But as the head goes from one side 

 to another, the relations of parts is much changed, though 

 quite gradually and automatically. It is this that makes 

 it difficult to give a description easy to understand. 



The actual line of the girth, at the middle of the move- 

 ment, when the larva is straight with the head aud legs 

 well raised, is behind the marginal process of the 1st 

 abdominal segment, then forwards above the marginal 

 processes of 2nd and 3rd thoracic, and then across the 

 larval venter to the other side in the incision between the 

 first and second pair of legs. I have said the head is well 

 raised, and so it is, by the Sphinx attitude of the first 

 segments, but it is strongly bent forwards, so that the tip 

 of the spinneret reaches very closely to the position of the 

 girth in the incision behind the first pair of legs. I say 

 very nearly, for the girth when completed consists of a 

 number of quite separate threads, showing that each 

 thread is not spun along, and glued to those that preceded 

 it, and that therefore the extremity of the spinneret does 

 not actually reach aud touch the previously spun threads, 

 which lie deeper in the incision between the segments. 



As the larva moves its head from side to side in adding 

 each thread, the position of the girth differs from this 

 central position by being stretched along one side and all 

 but relieved from the other ; when the head is round to the 

 left aud the left end of the thread is being fixed, the thoracic 

 and first two abdominal segments have their right sides 

 stretched so as to form the margin of nearly a circle, 

 whilst their left sides are so approximated as to be close 

 together at the centre of the circle. In this attitude the 

 line occupied by the girth above the marginal tubercles on 

 the stretched side of the larva is raised above the surface 

 on which the larva rests, and is on what for the moment is 

 rather the upper-side of the larva, though it would be 

 rather the under-side if the larva were in a natural resting 

 attitude, since, as I have above called attention to, the 

 larval warts are, and look, higher up on the larva than their 

 real anatomical positions would indicate. 



