114 Mr. G. C. Robson on 



the other hanci, we find localities wliere '^ intermediates " * 

 occur. Examination o£ these various types liave been 

 principally conchological, and it has been concluded that all 

 the shell-characters are completely '^ transgressive '"^ *. A 

 more limited examination o£ the genitalia indicates that these 

 organs are also transgressive. An exception has been invari- 

 ably made in the case of the dart. It is usually stated that 

 the latter is the only criterion of distinctness. There now 

 appears to be doubt even on this subject. 



We have also cases where the hoiHen sis find nemoralis com- 

 plexes are modified so as to render distinction impossible. 

 Information is less plentiful about the relation of these 

 '^species " to the other Tacheas. But it is possible that they 

 have the same indeterminate relationship with T. austriaca. 



Before compltting our review of this case we must briefly 

 consider the dart as a criterion of distinctness. As we have 

 pointed out above, there is some doubt whether it is not also 

 transgressive. A new difficulty is, however, created by the 

 discovery by Lang (4), 'Kleiner (3), and Boettger (2) that 

 the dart itself behaves not as a unit but as a complex. Eight 

 characters can easily be distinguished in it, and, according to 

 the results of Lang and Kleiner^ the darts of the Fj hybrids 

 are a ''mosaic^' of Itorteiisis, nemoi-aUs, and intermediate 

 characters ! Furthermore, all the hybrids did not exhibit the 

 same combination. 



We thus see that in this case the search for an absolute 

 criterion breaks down. Oar gene-complex appears to be a 

 constant association only in certain cases. At one time we 

 find an association of extreme Jiortensis or nemoralis 

 characters ; and we might be justified in recognizing such 

 compact groups of chaiacters as ^^ species." But in other 

 cases we find groupings which we can only assign to one or the 

 other of the two extreme types by an ohw'iou^ petilio principii — 

 an appeal to the dart. In assigning such forms to one or 

 another of the two "species " we are certainly not indicating 

 their real relationship, which is one that cannot be expressed 

 by crudely contrasted names. What we require is some 

 system that expresses the difierent modes of character- 

 combination. 



It is very much to be regretted that up to the present no 

 satisfactory Fo results have been obtained from crossing the 

 extreme forms. If an increased knowledge of the necessary 

 vital conditions is successful in producing a third generation, 



* The terms " intermediate " and '^ transgressive " have been used in 

 tliis note v^^ithout qualification. In the circumstfuices, a discussion of the 

 ])recise meaning of such types of variation seems to be unnecessary. 



