BOYCOn : r.OCAL VARIATION OF CLAUSIIJA BIDENTATA. 21 



priate methods, and it has, I hope, been shown that this sort of study 

 is readily practicable for the investigation of variation by working 

 field naturalists. It may want an actual trial to convince others that 

 it is as entertaining as I have found it. 



The demonstration of these differences in size is a preliminary to 

 the really difificult enquiry whether we can correlate thefn with any 

 characters of the individual loci. 



Arranging the data in descending order of altitude, we notice at 



Mean ilean 



Locus. altRiide. Character. diameter. 



H ii'473 roadside: moderate shade 2"6i7 



A 1 1436 roadside: moderate shade 2581 



I 1 1 429 roadside: open 2540 



J ii'348 roadside: open 2-566 



F II 228 roadside: some shade 2-559 



B 11091 roadside: open and exposed ^ 2*559 



I) 11024 tramroad bridge : sheltered, high ... 2'568 



G 10*489 densely shaded in wood 2-565 



C io'24i tramroad bridge : open, exposed, high 2-463 



E 10-171 densely shaded in wood 2-584 



once that three lots, G, C, and E, are conspicuously shorter than the 

 rest, and that of these C is also much narrower. C is certainly the 

 locus of the most inhospitable character by our human standards — 

 high lying, exposed, and bleak. ^ G and E, on the other hand, are 

 the very opposite — low, sheltered, damp, densely shaded places ; the 

 shells differ from C in being fatter. D is an old tramroad bridge, 

 very like C and not far from it but rouch more sheltered. At the 

 other end of the table, H and A are roadside walls with a consider- 

 able amount of shade, more shaded than F, I, or J, but much less 

 shaded than E and G. Putting all the facts together, it seems that 

 exposed windy places and densely shaded places produce short shells, 

 while moderately shaded loci are most favourable for the largest 

 growth ; roadside walls seem, too, more favourable than others.' 

 Why this should be so, requires of course much further analysis of 

 the cecological features along a variety of different lines ; this pre- 

 liminary enquiry enables one, however, to limit such an investigation 

 within practical bounds and examine a few loci, say H, C, and E, in 

 elaborate detail. Note, too, that these tentative conclusions as to the 



1 By '" open " I mean open to the sun, by " e.vposed " open to the wind and weather. 



2 In all the loci the substratum is the same sort of stone wall, and the only differences noted 

 were in its surroundings. 



3 Perhaps because of the dust which gets the mineral matter into a more readily soluble 

 slate for plants and animals icf. Science Gossip, vol. iv (1898), p. 343). 



