1 1 o RETROGRESSION 



of retrogressive variations, and abundance of nutriment a cause of 

 progressive variations. In other words, it is supposed that varia- 

 tions are not always spontaneous, but are, very often at least, 

 caused by the direct action of the environment a notion which, 

 in the main, Weismann himself has strenuously controverted. But, 

 if determinants supposing they exist as discrete elements in the 

 germ-plasm compete for nourishment, the competition should be 

 sharpest and most destructive when nutriment is least abundant, 

 and non-existent when it is superabundant. In that case species 

 which possess very abundant nutritive supplies (e.g. internal 

 parasites, unicellular and multicellular) should display little or no 

 retrogression, whereas species, the nutritive supply of which is so 

 small that they are engaged in an endless struggle for it, in which 

 many perish (e.g. most, if not all fishes, reptiles, birds, and 

 mammals) should display little progression and much retro- 

 gression. The exact opposite is the truth. The microbes of 

 syphilis remain microscopical though their determinants have 

 absolutely soaked in nutriment for thousands of years, and of all 

 known organisms the multicellular pasasites display the greatest 

 amount of retrogression. Some of them are little more than 

 egg-cases. Fishes, reptiles, birds, and mammals, on the other 

 hand, are typical examples of progression. In neither case has 

 the inevitable drift ensued which would necessarily have followed 

 were this hypothesis true a drift which would tend to abolish 

 Natural Selection, and therefore adaptation. 



1 8 1. Of all the splendid services rendered by Weismann to 

 biology none is more valuable than his long and successful 

 struggle to overthrow the Lamarckian doctrine. Since merely 

 useless parts tend to become vestigial and disappear, his success 

 involves the corollary that simple cessation of selection (i.e. cessa- 

 tion of utility), which may, or may not, imply cessation of use, is 

 followed by retrogression, which in turn implies that to prevent 

 retrogression a certain stringency of selection is necessary, while 

 to cause progression a still greater degree of stringency is required. 

 The retrogression of old-established parts is very slow. Thus the 

 great changes in whales and snakes since their limbs began to 

 retrogress indicates an enormous lapse of time. Doubtless, also, 

 geological epochs must pass before Greenland whales and pythons 

 lose even the small vestiges of hind limbs which they still possess. 

 On the other hand, newly acquired characters tend to disappear 

 very rapidly on cessation of selection, as may be noted in the 

 case of the special traits of our prize breeds. Absolutely new 





