NATURAL SELECTION 197 



(4) in one case the observations involve merely a com- 



parison between the variation of the natural popula- 

 tion (a) over a single season, and (b) over a period 

 of years (Kellogg and Bell, 1904). 



(5) One case related to the survival or death of animals 



brought into laboratory conditions after a pre- 

 liminary exposure to a generalised eliminating 

 factor, though the actual causes of death were not 

 controlled (Bumpus, 1899). 



(6) In three cases a special procedure was adopted, viz. that 



of comparing the variation of juvenile stages with 

 adult (Weldon, 1901, 1904 ; di Cesnola, 1907). 



(1) Weldon (i8gg). 



These experiments and observations are so well known that 

 they do not need to be explained in detail. Series of measure- 

 ments made by Weldon and his collaborator Thompson over 

 the years 1 893-1 898 on the crab Carcinus maenas in Plymouth 

 Sound showed that the mean frontal width of the carapace 

 (M.F.W. ) (expressed as a proportion of the length of the carapace 

 taken as = 1000) decreased in crabs of a similar carapace 

 length. Weldon attributed this to the elimination of crabs of 

 high M.F.W. through the action of silt in the gill-chamber 

 clogging the gills. He stated that the amount of silt in the 

 Sound had increased owing to the building of a breakwater 

 which prevented the escape of the detritus from china-clay 

 workings which was being washed into the Sound. Experi- 

 mental controls showed the following confirmatory results : 

 (i) Crabs were placed in vessels containing clay silt in suspen- 

 sion. Those that died had M.F.W. larger than that of the 

 survivors, (ii) Small crabs were collected on the shore and 

 kept in clean water. Some died — (?) from the effect of putrid 

 food. After the first moult the survivors were killed and 

 measured, and it was found that they were broader than wild 

 crabs of a similar size — which, on Weldon's hypothesis, is what 

 one would expect in the silt-free conditions. 



This work has been criticised by Cunningham (1928, 

 summary), Vernon (1903), Pearl (191 7), and Robson (1928). 

 The criticism falls into three categories : (a) as to the external 

 conditions ; (b) as to Weldon's assumption concerning the 

 relation between M.F.W. and filtration of the gill-chamber ; 



