438 



SCIENCE. 



[N. S. Vol. XX. No. 509. 



theory of development by discontinuous vari- 

 ations, or mutations as they are now called: 



We are -without doubt confronted in the ants 

 with the disappearance of this protective envelope 

 [the cocoon] which has become unnecessary 

 through the incessant care lavished on the progeny 

 by the workers. 



From the point of view of the evolution of in- 

 stinct it is interesting to observe that this disap- 

 pearance does not take place gradually by evanes- 

 cence of the cocoon, but suddenly, so to speak, 

 since larvae which are similar to one another and 

 produce similar adults, either make an absolutely 

 complete cocoon, without any signs of reduction, 

 or make none at all. 



This fact may be cited to show how sudden may 

 be the changes that supervene in the habits of an 

 animal. It supports the conclusions announced in 

 a communication by my brother, M. Armand Janet, 

 to the congress at Leyden. These conclusions, de- 

 duced from considerations of rational mechanics 

 applied to the problem of species regarded as a 

 position of equilibrium, tend to demonstrate that 

 tlie differences between a certain form and its de- 

 scendants are produced rather by quite sudden 

 leaps than by insensible and continuous variations. 



Much has been published of late, both by 

 the government and the newspapers, concern- 

 ing the advisability and feasibility of estab- 

 lishing the kelep ant in Texas and the other 

 cotton-growing states for the purpose of de- 

 stroying the boll-weevil. I feel at liberty to 

 comnient on this subject because for some 

 years past I have had rather exceptional op- 

 portunities in Texas, Florida, Mexico and the 

 Bahamas of studying the habits of a number 

 of species belonging to the same natural sub- 

 family as the kelep. This study has con- 

 vinced me that the attempt to establish the 

 ant in Texas will prove to be about as success- 

 ful and profitable as an attempt to acclimatize 

 in the same state some rare Central American 

 orchid, the South African secretary bird or 

 the Australian wombat. But in undertaking 

 to give my reasons for this opinion, I am far 

 from wishing that the experiment were not 

 being tried. An experiment may, of course, 

 be a complete failure from a purely economic 

 standpoint and still be of considerable, albeit 

 negative, scientific value. 



At the outset I may say that I have not my- 

 self studied the living Ectatomma tiibercu- 



latu'iUj though I am familiar with the insect in 

 collections.* But Dr. Cook's accountf shows 

 that in all essential particulars its habits are 

 the same as those of other Ponerina.:): 



The peculiarities which the kelep shares 

 with other Ponerinse and which would seem to 

 place serious obstacles in the way of establish- 

 ing it in a foreign country, are the following: 



1. The PonerinsB are archaic ants which 

 form small colonies§ and comprise no dom- 

 inant species except in Australia, where the 

 genus Myrmecia (the famous ' bull-dog ants ') 

 offers an interesting and instructive parallel 

 to the marsupialia among mammals. In fact> 

 the opossums of America bear about the same 

 relation to our dominant mammals, like the 

 rodents, as do the Ponerinse to the dominant 

 ants of the subfamilies Myrmicinas and 

 Camponotinffi. 



2. The Ponerinse are local and rare, and 

 show little adaptability or plasticity in com- 

 parison with the more recently evolved and 

 dominant species. 



3. It is altogether exceptional for any of the 

 Ponerinie to restrict their diet to a single 

 species of prey. In fact, the only known 

 exception is the Texan Lepiogenys elongata, 

 which, as I have shown, feeds very largely, 

 if not exclusively, on terrestrial isopods.^f The 

 kelep is described by Dr. Cook as capturing 

 and consuraing ' adult insects of many and 

 diverse kinds,' so that we may be sure that it 



* Dr. Cook is in error in supposing that there is 

 anything doubtful about the occurrence of E. 

 tuheroulatum in Mexico. I have before me a fine 

 series of workers taken at Tuxpan by Mr. J. F. 

 McClendon. They agree perfectly with a typical 

 specimen of the kelep collected by Stoll, at 

 Retalhuleu, Guatemala, and sent me by Professor 

 Forel. 



t ' Report on the Habits of the Kelep, or Guate- 

 malan Cotton-Boll-Weevil Ant,' U. S. Dept. Agr. 

 Bur. Bull. No. 49, Washington, 1904. 



+ See my papers, ' A Study of Some Texan 

 Ponerina;,' Biol. Bull., Vol. II., No. 1, 1900, and 

 ' The Habits of Ponera and Stigmatomma,' ibid., 

 Vol. II., No. 2, 1900. 



§ £. tuberculatum colonies contain from 20 to 

 110 workers each, and average between 40 and 50. 



Tf' A Crustacean-eating Ant {Leptogenys elon- 

 gata Buckley),' Biol. Bull., Vol. VI., 1904. . 



