176 [August, 



and if more specimens were examined would possibly prove to be 

 unimportant. But as regards the wings the reverse is the case. 



H. ytenensis abounds at Brockenhurst, and when I found, a year i 

 ago, an individual with the wings distinctly longer than usual (tig. 75), I 

 I suspected that by examination of a large number of specimens Ij 

 should find a series of intermediates. I have therefore captured andj 

 examined some lunidreds of specimens of ytenensis, but have found no 

 real connection with gramdaris in this respect. The wings are not 

 only shorter than in (jrannlaris, but are differently shaped. I figure! 

 the wing of grannJaris (tig. 73), which is the same in all the examples 

 of that species (24 in number) at my disposal, and the wings of the ' 

 various forms of ytenensis selected from a vei'y long series: fig. 74 j 

 shows the form as it usually occurs ; fig. 7Q a very rare, broad variation 

 in shape ; and fig. 75 the longest wing of a series of some hundreds 

 of examples. These figin-es show that the wing of ytenensis is some- 

 what unstable in size and form. The difference in size is not correlative 

 with that of the individual. I find it impossible at present to account 

 for the wing distinctions between granuJaris and ytenensis by any theory 

 except that the two are really different species. 



At Brockenhurst the two forms occur together, and about 3 per 

 cent, of the total specimens captured have proved to be granidaris ; 

 the two, however, are not mixed indiscriminately, for the spot that has 

 produced most of the specimens of granidaris is one where ytenensis 

 is rare, and the other localities where ytenensis is abiindant at present 

 have each yielded only one or two granidaris. 



The species inhabits very shallow grassy ditches or pools, that dry 

 up completely for four or five months each year, and are then covered 

 with ordinary, as distinguished from aquatic, vegetation. In Novem- 

 ber, and in the spring from March till April or May, ytenensis may be 

 found in profusion in certain of these spots, and has continued to 

 frequent tlie same spots for several sriccessive years. The place where 

 granidaris occurs is a larger pond with aquatic vegetation, and only 

 dries imperfectly and leaves much mud in the summer. This jiond is 

 only a few yards distant from a shallow ditch in which ytenensis 

 abounds, so that the few specimens of this species found in the pond 

 may be merely temporary migrants. A corresponding consideration 

 might account for the occasional presence of a specimen or two of 

 grannlaris in the haunts of ytenensis, so that the evidence inclines one 

 to believe that there is some real difference in the habits of the two 

 forms. 



