A COMMENTARY ON LAMARCKISM 



the same period, and the untrained controls improved from 

 149 to 102 over a period of only four years. 



Careful independent repetitions of McDougalPs work by 

 Crew (1936) and Agar, Drummond and Tiegs (1935, 1948) 

 failed altogether to confirm his empirical findings; they were 

 not scrupulously exact repetitions, it is true, but embodied 

 refinements that increased the precision of the experiments 

 without in any way aff'ecting the principle of their design.* 

 McDougalPs results are therefore on a somewhat different 

 footing from those of Guyer and Smith. ^ What part could 

 selection have played? In theory no part, for the experimental 

 subjects had been inbred for a sufl^icient number of generations 

 to justify the prevailing theoretical assumption that they were 

 genetically uniform and homozygous. In practice, this pre- 

 sumption seems to have been unduly optimistic: Loeb"'s work 

 (1945) on the transplantation of tissues between members of 

 the highly inbred Wistar strain of rats revealed incompati- 

 bilities that can only have been due to flagrant heterozygosity. 

 Guinea-pigs and mice, by contrast, become completely tolerant 

 of grafts transplanted between members of an inbred line after 

 a much less prolonged regimen of inbreeding. McDougalPs 

 stock may, then, have been more heterogeneous than is usually 

 supposed — and, as Drew (1939) has made clear in his admirably 

 succinct review, there is plenty of evidence that differences of 

 intelligence in rats, as measured by maze performances, are 

 perfectly amenable to selection. 



It may of course be argued that McDougalPs adverse selec- 



1 Haldane (1951) makes the comment that McDougall's colleague and 

 pupil, Rhine, was conducting experiments in paranormal cognition in the 

 same laboratory, and points out the inconsistency of presenting evidence 

 in favour of paranormal cognition in human beings without taking into 

 account its effect on the outcome of such experiments as McDougall's. 



* [The final report on this long and important experiment has now been 

 published: W. E. Agar, F. H. Drummond, O. W. Tiegs and M. M. Gunson, 

 Journal of Experimental Biology, 31, p. 307, 1954.] 



95 



