50 LIFE : OUTLINES OF GENERAL BIOLOGY 



ing of fleece after transport to a cold country, or the gradual blinding 

 of goldfishes in a dark room. Sometimes the modification lasts for 

 the rest of life; sometimes it is gradually lost. Modifications take a 

 firmer grip than do transient adjustments, but some do not last a 

 lifetime as others do. 



(5) In many cases there is reason to believe that the environment 

 affects the organism for better or for worse, although there is 

 nothing to show for it. We mean that it may affect the germ-cells 

 without appreciably affecting the body or soma. This is indicated 

 by many experiments. Some of these show that the general vigour 

 of the germ-cells may be improved or depreciated ; and in experi- 

 mental conditions it may be that the environment, e.g. through 

 radium rays, kills the germ-cells. On the other hand, ultra-violet 

 rays may improve the whole vigour of the organism, germ-cells 

 included. Where viviparity occurs, and especially in cases of ante- 



Variations in Potato Hectic {Leptinotarsa decemUneata). After Tower. The 

 three forms (1-3) differ in the details of their marking. 



natal symbiosis, as in placental mammals, the vigour of the 

 offspring may be increased or decreased by the quality of the 

 surroundings. 



(6) Hy themselves we are inclined to place those cases where a 

 change in the environment of the parent provokes a variation in 

 the offspring, through some disturbance in the germ-cells. Thus 

 Tower subjected potato- beetles (Leptinotarsa) to unusual conditions 

 of temperature, pressure, and humidity, when the male or female 

 reprcKluctive organs were at a certain stage of development. The 

 body of the parents showed no modification, but there were in some 

 cases remarkable changes in the offspring, in colour and markings, 

 and even in minute details of structure. There was no reversion in 

 the offspring of these variants. Here the environment acted as a 

 stimulus to variability, and it is likely enough that this often occurs 

 in Nature. Hut a repetition of Tower's important experiments is 

 urgently needed. 



(7) Quite different from the preceding relations is the selective 



