D. J. SCOURFIELD OX MENDELI8M AX!) MICROSCOPY. 417 



segregation of characters), and some of the larger Rotifers, such 

 as Asplaachna and Hydatina, might very well be considered in 

 this connection. The Hydrachnids, which at first sight seem 

 offer some possibilities from this point of view, are, 1 am afraid, 

 ruled out of court by the fact that they pass through a peculiar 

 series of active and resting stages, rendering it impossible to re 

 them to the adult form. If points of difference could be made 

 out in the larvae of nearly related forms, however, somethi 

 might be accomplished even here. 



The principal aim of this paper has now been accomplished, for 

 we have given some consideration to Mendel's own work, to the 

 recent developments on the same lines, and to the connection of 

 the ideas so evolved with microscopical investigations. But 

 before concluding I would like to point out quite briefly the bear- 

 ings of Mendelism on some of the : current notions about heredity, 

 variation, and evolution. 



According to a very prevalent conception, a "pure" race can 

 only be one whose ancestors for a great many generations have 

 never been crossed with any other race, and have never been 

 known to produce characters other than those peculiar to the 

 race. According to Mendel's conception of gametic purity, how- 

 ever, it is evident that, so far at least as certain characters are 

 concerned, a pure race can originate at any point, even in the 

 first generation from hybrids. The one thing essential for the 

 production of a pure race is that germ-cells bearing only factors for 

 similar characters should unite, and this, as Mendel showed, can 

 take place when hybrids are interbred as well as when pure forms 

 are mated. So far, then, from each organism necessarily trans- 

 mitting some influence from all its ancestors for many genera- 

 tions back, as would be the case if the commonly received ilea 

 were correct, it may not in some cases even transmit characters 

 which appeared in one of its own parents. The practical impor- 

 tance of this in connection with the rapid fixing of useful 

 characters is self-evident. 



So far as variation is concerned, Mendel's law certainly sup- 

 ports the idea that this is a term covering many different classes 

 of facts. In the first place, the acceptance of the law leads 

 directly to a still more strenuous insistence upon the distinction 

 between variations due to alterations in the germ-plasm and thi 

 due to the reaction of the body of an organism to its environment, 



