1879.] on the Ilahits of Ants. 187 



not more than would naturally have resulted from the number of ants 

 running about. Wo then, at 3, allowed the ants which were feeding 

 to return to the nest. In less tlian half an hour after this eleven 

 came, and in the following half hour no less than forty-three. So 

 that in the first four hours only seven came, while in less than an hour 

 after the first was allowed to return to the nest, no less than fifty-four 

 came. 



Again, on September 30th, I tried the same arrangement, beginning 

 at 11 ; up to 3 .30, seven ants came. We then let them go. From 

 3.30 to 4.30, twenty-eight came. From 4.30 to 5, fifty-one came. 

 Thus, in four hours and a half only seven came ; while when they 

 were allowed to retui-n, no less than seventy-nine came in an hour and 

 a half. 



I tried this experiment several times more, and always with 

 similar results. It seems obvious, therefore, that, in three cases at 

 least, no communication was transmitted by sound. 



I will now endeavour to show you one or two microscopical pre- 

 parations, merely to give you a very slight idea how beautiful and 

 complex the anatomical structure of an ant is. Here is a longitudinal 

 section of a queen ant. 



Sir John then proceeded to describe the principal points : — 

 The organs of vision of ants are generally well developed and 

 conspicuous. There are usually three simple eyes, or ocelli, arranged 

 in a triangle on top of the head, and on each side a large compound 

 eye. These compoimd eyes are very complex organs, but the mode in 

 which they act is by no means understood. They consist of a number 

 of facets, varying from 1-5 in Ponera contracta, to more than 1000 

 in each eye — as, for instance, in the males of F. pratensis. In fact, 

 these, so far fortunate, insects realize the wish of the poet — 



" Thou lookest on the stars, my love, 

 Ah, would that I could be 

 Yon starry skies, with thousand eyes 

 That I might look on thee." 



But if the male of F.jpratensis sees 1000 queens at once, even when 

 only one is really present, this would seem to be a bewildering j)rivi- 

 lege, and the prevailing opinion among entomologists is that each 

 facet only takes in a portion of the object. 



From the observations of Sprengel, there could, of course, be 

 little, if any, doubt, that bees are capable of distinguishing colours : 

 but I have, in my previous lecture, recorded some experiments which 

 put the matter beyond a doubt. Under these circumstances, I was 

 naturally anxious to ascertain if possible, whether the same is the 

 case with ants. I have, however, experienced more difficulty in doing 

 so, because ants find their food so much more by smell than by sight. 



I tried, for instance, placing food at the bottom of a pillar of 

 coloured paper, and then moving both the pillar and the food. The 

 pillar, however, did not seem to help the ant at all to find her way to 



o 2 



