February 23, 1906.] 



SCIENCE. 



309 



tioH, it will produce a great number of indi- 

 viduals forming what we call a common and 

 polymorphic species, probably more or less 

 variable in habits, or polytropic. When the 

 number of individuals approaches the maxi- 

 mum, the pressure of competition increases 

 until the position becomes unfavorable to a 

 certain proportion of the individuals. The 

 pressure is least on the set of individuals hav- 

 ing the most divergent habits. There is a 

 tendency for the polymorphous polytropic 

 group to break up into a number of more 

 uniform oligotropic groups. If the parent 

 species is itself oligotropic, it may give rise 

 to a form of quite different habits. The orig- 

 inal form retains the original position and 

 the derived form changes to a new position. 

 Usually all that is required to place a form in 

 an absolutely new ecological position and 

 make it the progenitor of a varied line of 

 descendants is a mere change in the kind of 

 food. The assemblage of bees owes its exist- 

 ence to the fact that some aculeate hymen- 

 opteron abandoned the pursuit of other in- 

 sects and provisioned its nests with honey and 

 pollen. 



Whenever competition becomes severe, nat- 

 ural selection may operate upon two sets of 

 individuals, those which have the original 

 habits and those which show a change of 

 habits. In the first set it retains the most 

 perfectly adapted individuals. This merely 

 keeps the original species adjusted to the 

 original habits. The second set becomes the 

 new species and natural selection may further 

 operate to fit it to the new habits. When an 

 old organ is used for a new purpose, we can 

 understand how, after competition has again 

 become severe, individuals in which the organ 

 is best fitted for the new use will have the 

 advantage. The theory of natural selection 

 itself as applied to adaptive structures implies 

 that the development of a new organ is pre- 

 ceded by a change of habits, for how is natural 

 selection to improve an organ for a certain 

 purpose unless the organ is already being used 

 for that purpose? 



Those who attach so much importance to 

 structural modifications as a condition of 

 selection seem to me to overlook an important 



element in the nature of adaptation. Adapta- 

 tion is determined by the nature of the posi- 

 tion as well as by the presence of structures 

 fitting the organism for it. The bird which 

 became the progenitor of the humming birds 

 was better adapted to its new place than any 

 of its more modified descendants, not because 

 it was structurally better fitted to get nectar 

 from flowers, but because it occupied a more 

 favorable place. The favorable nature of this 

 place is shown in the fact that it could pro- 

 duce 400 specific forms in a comparatively 

 short time. The absence of adaptive struc- 

 tures does not show that a species is wanting 

 in adaptive habits of the most distinct kind. 

 It does not show that natural selection has 

 been any less operative in producing segre- 

 gation. 



As an example of ecological adaptation I 

 may mention several species of the bee genus 

 Colletes which occur in my neighborhood. 

 They are distinguished by structural differ- 

 ences of the labrum, antennse, metathorax, by 

 size and punctation and by the color and 

 arrangement of the hair. The proboscis and 

 pollen-carrying apparatus are the same in all, 

 and, as far as I can see, the species do not 

 have any adaptive structural differences. The 

 nine species so differ in habits and in seasonal 

 distribution that only three are in competi- 

 tion. C. inwqualis is polytropic and flies 

 from March 20 to May 31. C. cestivalis gets 

 its pollen exclusively from Heuchera hispida 

 and flies from May 8 to July 1. C Irevicornis 

 is an oligotropic visitor of Specularia per- 

 foliata and flies from May 29 to June 29. 

 G. willistonii flies from May 28 to September 

 5, and C. latitarsis from June 13 to September 

 29. Both of these are competitors for the 

 pollen of Physalis, but they are not closely 

 related. 0. latitarsis is more common and its 

 flight begins and ends later. C. eulophi flies 

 from May 28 to October 9. It is polytropic, 

 but gets most of its pollen from Compositae. 

 It is a competitor of the following species 

 only in 40 per cent, of its days and in 22 

 per cent, of its observed flower visits. Three 

 species are autumnal and get their pollen 

 exclusively from Compositae: 0. armaiiis, 

 August 17 to October 7; C. americanus, 



