594 PROFESSOR MARCUS HARTOG. 



the scent of the rose still clings to the emptied vessel ; (2) it 

 will not conveniently yield an adjective to apply to its modes, 

 etc. This latter objection has been tacitly felt by most 

 writers, who have consequently used the terms " sex/' 

 "sexual/' etc. — extending this term, which originally implied 

 a binary differentiation, to all cognate phenomena, 

 whether there exist such differentiation or no. Thus my 

 friend Mr. Wager has written a most important and valuable 

 paper on the "Sexuality of the Fungi," though no differentia- 

 tion of male or female exists in some of the most important 

 and, indeed, primitive types. The word is the more un- 

 fortunate, for the user of this terminology is uuconsciousl}' 

 swayed by the implicit idea of such binary differentiation 

 into two contrasting categories of beings or cells as exists 

 among ourselves. Yet we have seen that in the most 

 primitive cases the fusing-cells are to all intents and purposes 

 identical — nay, more than two may fuse into a single cell. 

 In isogamy with exogamy, so common in Protista, any one 

 gamete will pair with any other, provided that it belong to 

 a different brood to its own. It has been suggested that here 

 we have a sort of foreshadowing of sexual differentiation, but 

 the suggestion will not hold water for a moment. Let us 

 consider twenty-six broods of pairing-cells matured at the 

 same time, and letter them with the letters of the alphabet, 

 and suppose that their exogamy be a glimmering of sex. 

 Then we may suppose that A is of the male sex, and that 

 with respect to it B, C, D . . . Z, are all more or less females ; 

 the same applies to B with respect of C ; and in the same 

 way we could show that any one brood is niide and female at 

 once — that is, that they are sexually undifferentiated. There- 

 fore, etc., Q, E, D. Again, in the Heliozoan Actinosphterium 

 the pairing-cells are second cousins by the laws of cellular 

 kinship, and have had precisely the same history from the 

 grandparent-cell. The suggestion that there can be any 

 binary differentiation in such cases has arisen sim])ly from 

 the associations inseparable from the word " sex/' and the 

 only ground for the assumption of latent differentiations 



