procession of Cnethocampa pinivora. 583 



In this there is little or no suggestion that the threads secreted 

 in the circulating mass help to preserve the order of a procession, 

 for even if the threads of individual larvae are secreted continuously 

 in a voluntary mass they are certainly broken when the mass is 

 made artificially. Edwards {loc. cit. p. 435) suggests that the formation 

 of a mass may be a means of altering the order of the larvae while 

 retaining the same leader, but there seems to be little evidence for 

 this view. 



Moreover the mass formation may continue for an hour or two 

 before a new procession is formed, and as threads continue to 

 be secreted the larvae are in a tangle of silk after a prolonged 

 massing, in which it seems impossible any individual thread can 

 be distinguished by them. It is true that the number of cases 

 in which the old leader was retained is, both in Edwards' and my 

 own cases, higher than mere chance would give, and a possible 

 explanation is that in any procession one larva has a greater 

 tendency to stray than the others, and this one is thus more likely 

 to act as leader. I found that it is not at all uncommon for one 

 larva to stray from a mass and to repeat this again and again 

 when put back artificially. Such specially erratic larvae secrete 

 a thread but it is disregarded by the others. These independent 

 larvae sometimes return to the mass voluntarily after an excursion 

 several inches away. In this case and also if put back artificially 

 they are often seen to relinquish the wandering and to take any 

 place in a new procession. It may be suggested that the pro- 

 cession is reformed from a mass by a larva with the wandering 

 habit relatively marked tending to leave it, while the other larvae, 

 as can always be seen when a procession forming is watched, in the 

 endeavour to retain contact become parallel with it, at the same 

 time falling behind because it walks faster. One of its fellows 

 thus soon finds its head in contact with the leader's tail, which 

 is the processional position. The same thing happens for all the 

 larvae in succession and these fall from two or three abreast into 

 single file till the new procession is fully arranged. Processions 

 are invariably formed by the larvae coming into head-and-tail 

 contact in succession. I have never seen them make head-and-tail 

 contact simultaneously with their bodies bent and then the 

 procession gradually straighten. What happens is that the pro- 

 cession emerges as a straight line from a continuously diminishing 

 circulating mass. 



The above observations suggest that the thread is of very little 

 or no value in the formation of a procession, and that its help 

 in maintaining the integrity of a procession is slight. It seems 

 to be secreted all the time that larvae are in active movement, 

 whether on the march, in mass or burying. It forms the nest 

 in the tree and the cocoon of the pupa state. Why its secretion 



