71 



Melitcea, or Colias would serve our purpose for illustrating 

 this point equally well, and it becomes necessary before 

 saying that the differences between species are speciiic to 

 have a deiinition of the term species, an almost impossible 

 thing in the present state of our knowledge, for the defini- 

 tion is little more than a matter of personal opinion in all 

 cases of real difficulty. Up to a certain point it is true to 

 say that the differences between the various Anthrocerids 

 just named are differences of kind, but do these differences 

 form a really discontinuous series ? One may also ask 

 whether the differences are not in reality differences of 

 degree rather than of kind, just as the diversities of environ- 

 ment are stated to be, on the whole, differences of degree ? 

 But do the differences of environment form a continuous 

 series? Is there not a distinct difference between the 

 environment of Anthrocera filipendiilce and Polyommatus 

 corydon that live on the same chalk hills in Kent ? Is 

 there not a distinct difference between the environment of 

 the meadows of Kent that produce Anthrocera trifolii and 

 the environment of the marshes at Freshwater that produce 

 A.palustris? There is a distinct climatic difference between 

 Algeria and Kent ; but this does not constitute the environ- 

 ment of an insect occurring in both districts, and therefore 

 I am not surprised when I cannot satisfactorily distinguish 

 between the meadow-haunting A. syracusia (as the Algerian 

 A. trifolii is called) and the specimens oi A. trifolii I take in 

 the meadows at Strood. Still I can understand that the 

 difference of latitude between Algeria and Kent may con- 

 stitute an important factor in the environment of a species 

 found in the two countries, and I am not surprised to find 

 that A. serisiati (the Algerian form oi A. palnstris) varies 

 distinctly from the A. palustris taken in the marshes of 

 Sandwich or the Isle of Wight. But the effect of environ- 

 ment is evident, for although we are unable definitely to 

 separate specifically the meadow-loving A . trifolii from the 

 marsh-haunting A. palustris, often to be taken within a few 

 miles of each other, as at Canterbury and Sandwich, yet the 

 meadows of Algeria produce the meadow form of A . trifolii, 

 and the marshes between Collo and Bono the form we are 

 so familiar with as palustris from the marshes of Fresh- 

 water, Sandwich, &c. That the actual difference observable 

 between English palustris and Algerian seriziati is due to a 

 peculiar factor in one or other of their respective environ- 

 ments we are willing to admit ; but we would urge that the 

 environment of A. trifolii in the meadows of Algeria and 



