100 Mr Brindley, The 'procession of Gnethocampa pinivora. 
procession returning loops about till it finds the outward bound 
thread ; (4) there is no special primite, and if a primite be 
removed artificially the next larva becomes leader with little 
hesitation ; (5) removal of a satellite and the thread under it 
results as a rule in the formation of two chains, each going its 
own way, while less frequently a single chain is formed by the 
two daughter chains joining up ; (6) there is no satisfactory clue 
to the influences which determine the course of the larvae, except 
that in some cases sunlight seems to attract the processions, and 
Fabre thinks it possible that warmth is really the stimulus. 
Gnethocampa pinivora is common in the Pinus Pinaster 
woods of the Landes, and on April 2, 1906, my wife and myself 
found a procession in the Cap Ferret woods, which separate the 
Gulf of Arcachon from the Atlantic. In the more open spaces 
between the masses of pines there are much sea holly and 
common yellow broom. The day was warm, with constant sunshine, 
and the procession proved to be the final one, i.e., for pupation. 
The number of larvae composing it was 114, and thus it was not 
a very long one, for Fabre states that he has seen a procession 
as long as 12m., containing about 300 larvae; and Mr Cleveland 
Morgan, Trinity College, informs me that in March, 1905 he 
found on the hills of the Algerian Tel processions of about 
180 larvae, the species in question being either C. pinivora or 
some closely allied form. (The nomenclature of the Eupterotidae 
is in a very confused state, and it appears that much uncertainty 
exists as regards their distribution and specific characters.) The 
procession observed at Cap Ferret was moving without pause 
over the undulations and low vegetation of the sand, the primite 
rotating his head from side to side with a suggestion of 
“ casting about.” It seemed clear, however, that the general 
direction taken was after the greatest sunlight, and this held 
also in the main for the daughter processions formed by inter- 
ference later on. That warmth, as Fabre has suggested, may 
be a guiding influence, and that in the more open and so more 
sunlit spaces there are wider expanses of sand free from 
vegetation and therefore perhaps more suitable for burrowing, 
are possibilities which suggest themselves in this connection. 
At the time of this observation I had not, to my subsequent 
regret, read Fabre’s work, and so the few experiments made 
lacked its valuable guidance. The chief of these were : 
(1) Placing every other larva of 30 contiguous ones on its 
own level about 15 cm. to one side of those left in position, thus : 
this resulted in all the larvae casting about ” for a time though 
