\ 



88 



THE BEGINNINGS OF IIFE. 



as to produce an organism similar to that from which 

 they were derived? It is well known that minute 

 fragments cut from a Hydra or from a Medusa~^\,^^ 

 placed in suitable situations— are capable of developing 

 into perfect organisms similar to those from which 

 they had been derived; just as a mere fragment of 

 a crystal thrown into an almost super-saturated solution 

 of the same salt will lead to the formation of a perfect 

 crystal. We are told, moreover, by Dr. Hooker ^, that 

 there is « a species of Begonia^ the stalks, leaves, and 

 other parts of which are superficially studded with 

 loosely attached cellular bodies,' and that ' any one of 

 these bodies, if placed under favourable conditions, will 

 produce a perfect plant similar to its parent.' The 

 power of repair and reproduction of lost parts which 

 exhibited by many animals comparatively high in 

 the scale of organization, is due to' precisely the same 

 causes. Multiplication by gemmation is, in fact, only 

 an extreme form of the phenomenon which takes 

 place when the crustacean reproduces a lost limb, 

 when the lizard and the triton reproduce a tail that 

 has been accidentally or purposely severed, or even 

 when the fish reproduces a similar part 2 : both sets of 



^ 'Report of Brit. Association/ 1868. 

 Facts of this kind have been only recently made known in respect 

 to animals so high in the scale as fish. The following quotation 

 CAthen^um/ Aug. 19, 1871) refers to a communication made at the 

 meeting of the British Association in Edinburgh : — ' Prof. Traquair 

 described two specimens of Protopterus annectens, in which the external 



configuration and internal structure rendered it evident that a consi- 



IS 





/ 



d sp«' 



Ifltieseco' 



U tip of ^' '''"' 



restorative procc 



undarybe 



iHd OB the outside 

 ite of tlie specimen' 



pileriorextremitj'ofthea 

 likas still proportiona' 



Tie restored porl 

 OE faection showed 



.but also 



Mliese eleme., 



iBitagmucliinoreirrq 

 ^^se to be traceabh 



^^Ptionofthetail.ilK 



tabtion. In> 



"' *' fa ^ys and U 



^ ^^ stales on the e 



""'^'"•arcatioabe 



I 



-ttfl. 



h. 



cit., 



P'l8i:op 



"•Si,, 



9s 



V 



k 



