CORRELATED VARIATIONS. 81 



agree with this conclusion, but thinks that the differ- 

 ences observed are too large to justify such an assump- 

 tion. 



At Professor Weldon's suggestion, Mr. E. Warren * 

 undertook similar measurements on 2300 specimens of 

 another crab, Portunus depurator, also obtained from 

 Plymouth. The accompanying table gives the results 

 obtained by Warren, and some of those obtained by 

 Weldon: 



C. MCENAS, PORTUNUS 



C. MOENAS, PLYMOUTH DEPURATOR, 



ORGANS. NAPLES RACK. RACE. PLYMOUTH. 



Total breadth and frontal breadth, .08 .10 .14 



R. antero-lateral, .66 .65 .67 



" " R. dentary margin, .50 .55 .56 



Frontal breadth and R. antero-lat., .29 .24 .30 



" " R. dentary margin, .23 .18 .03 



L. dentary margin, .26 .20 .01 



R. antero-lateral and L. antero-lat., .76 .78 .86 



R. dentary margin, .71 .78 .80 



" " L. dentary margin, .60 .70 .74 



On glancing through this table, it will be seen that, 

 with two exceptions, the values for the two races of 

 C. mcenas differ from one another nearly as much as 

 they do from the constants of Portunus, an animal be- 

 longing to a different genus. It is probable, however, 

 that the larger differences in this latter animal do indi- 

 cate real differences in the correlation constants, asso- 

 ciated, perhaps, with changes in habit or environment. 

 For example, it is conceivable that a crab which swims 

 might find it advantageous to be more symmetrical than 

 one which only crawls between the tide marks. Por- 

 tunus does swim to a certain extent, and we see that 

 the correlation of the two sides of its body is greater 

 than in the essentially shore-living Carcinus. 

 *Proc. Roy. Soc., Ix. p. 221. 



