ALTERNATION OF GENERATIONS 481 



regular succession, this phenomenon is that which is now understood 

 as the normal Alternation of Generations in Plants. 



It seems probable in many, though not perhaps in all phyla, that 

 nuclear fusion and reduction remained constant features in each 

 completed life-cycle throughout Descent. In that case two oppor- 

 tunities for somatic amplification were possible in Evolution from 

 simpler forms. The one between syngamy and reduction would 

 give rise to the body of tissue called the sporophyte. The other 

 between reduction and syngamy would give rise to the gametophyte. 

 If the fusion and reduction retained their identity throughout descent, 

 these two somata can never have been homologous, that is, homo- 

 genetic by descent. They must have been " antithetic " throughout, 

 however nearly they may resemble one another in their characters. 

 If they both develop in the same medium, they being in fact 

 merely phases of the same organism might be expected to resemble 

 one another very closely. That is found to be actually the case in 

 many Algae and Fungi. For instance, in Dictyota among the Brown 

 Algae, and Polysiphonia among the Red, the two generations 

 appear identical ; they seem only to differ in their chromosome- 

 number, and in the propagative organs which they bear. Similarly 

 in the Uredinae (p. 449), the paired nuclei of the sporophyte 

 on the Wheat are contained in hyphae similar in their main 

 features to those of the gametophyte on the Barberry, which have 

 only single nuclei. Much the same is the case in those Ascomycetes 

 in which evidence of alternation may be traced. There 'may be a 

 difference of potentiality between the haploid and diploid phases ; 

 but it need not be realised where the circumstances are uniform, as 

 when both grow in water, or in moist conditions. In that case both 

 may appear alike. 



ALTERNATION IN THALLOPHYTES. 



The cytological data for the Thallophytes are still so imperfectly known that 

 all comprehensive statements as to alternation in them must be provisional. 

 They probably represent a large number of phyletic lines. In those which 

 show sexuality the life-cycle appears to have been subject to the twc alter- 

 nating events of syngamy and reduction, but to have been worked out in 

 various ways in respect of somatic development in relation to them. A few 

 examples will be cited as showing how various the results have been. A very 

 simple but instructive case is that of the Diatoms and Desmids. In the 

 Diatom, Rhopalodia, it has been shown that a tetrad -division of nuclei 

 precedes conjugation. In the Desmid, Closterium, a tetrad-division succeeds 

 conjugation. If, as is probable, reduction is the consequence of that 



B.B. 2H 



