342 THE CELL 



similar molecules of albumen, since the portions obtained by 

 dividing a molecule, consist of groups of atoms of unequal value. 

 On this account idioblasts are not identical with the plastidules, 

 the existence of which is assumed by Elsberg and Haeckel (IX. 8 b). 

 For, according to Haeckel, the latter possess all the physical pro- 

 perties, which physicists ascribe to molecules, or to collections of 

 atoms, in addition to especial attributes, which belong exclusively 

 to themselves, viz. " the vital properties which distinguish the 

 living from the dead, and the organic from the inorganic." 



Our units, therefore, the gemmulae of Darwin, the pangenae of 

 de Vries, and the physiological units of Spencer, must be complex 

 units, or, at any rate, groups of molecules. In this fundamental 

 view, all the above-mentioned scientists agree. Thus, according 

 to Spencer, there is nothing left but to assume, that chemical 

 units combine together to form units of an infinitely more 

 complex nature than their own, complex though this be, and 

 that in every organism the physiological units, produced by 

 such combinations of highly complex molecules, possess various 

 characters." 



If Niigeli's hypothesis of the molecular structure of organised 

 bodies be accepted, it is easy to imagine that the nature of the 

 idioblasts is as follows : " They can as little be single micellae 

 (crystalline molecule-groups), as molecules; for even if, as a 

 mixture of different modifications of albuminates, they possess 

 different properties, they would still lack the capacity of multi- 

 plying and forming new similar micellae. Insoluble and stable 

 groups of albuminous micellae alone afford all the necessary 

 conditions for the construction of the gemmulse ; they alone, in 

 consequence of their Varying composition, can acquire all the 

 necessary properties, growing indefinitely by storing up micellae, 

 or multiplying by means of disintegration. Hence, the pangenae 

 or gemmulae must consist of small masses of idioplasm." 



Now comes the question : What is the size and number of 

 the idioblasts contained in a complete germ ? 



As regards size, the idioblasts must certainly be exceedingly small, 

 since all the hereditary elemental germs of a highly-developed 

 organism must be present in the minute spermatozoon. Nageli 

 has attempted to make an approximate calculation on this impor- 

 tant point. He starts with the assumption, that the hypothetical 

 albumen formula of chemists, with seventy-two atoms of carbon 

 (C 72 H 106 M 18 SOo 2 ), does not represent a molecule of albumen, but a 



