862 TRANSACTIONS AND PROCEEDINGS OF [Skss. lvi. 



on the nucellar cells surrounding it is shown in an early 

 and in two advanced stages in ligs. 15, 27^ 30. 



The next step in the life-history of the cell is its division, 

 and I regret to say that, amongst the many thousands of 

 sections I have made throuo-h flowers of all ages, I have not 

 found a single ovule throwing light upon the way in which 

 this division takes place. Erom Strasburger's brilliant 

 researches we know that indirect division occurs in other 

 plants, but how it takes place in Mi/osurns, and what becomes 

 of the vacuoles during the time of division I am unable to 

 state. Strasburger s.tates that no division of the protoplasm 

 occurs after the division of the nucleus of the sporocyte 

 (Strasburger's primary nucleus of the embryo-sac-cell), and 

 therefore concludes that the two newly formed nuclei can 

 not be regarded as the nuclei of two sporocytes (Warming- 

 Vesque view), but that they must belong to one individual 

 structure, namely, the female prothallus of a macrospore ; in 

 other words, that the cell I have called the physiological 

 sporocyte is a spore giving rise to a female prothallus, and is 

 not the e(|uivalent of two sporocytes. Already it has been 

 pointed out that the embryo-sac-cell must be a sporocyte, 

 because of the gelatinisation occurring in its walls, and because 

 of the nucleus of the embryo-sac-cell being compai-able to 

 the nucleus of a male sporocyte according to Guignard's 

 observations, and a third reason will become apparent as 

 we trace the further liistory of the two nuclei. 



The earliest stage I have found after the division of the 

 nucleus of the physiological sporocyte (Strasburger's primary 

 nucleus of the embryo-sac) is represented in fig. 16, showing 

 a sac with an apical and a basal nucleus surrounded by pro- 

 toplasm and a large vacuole in the centre of the protoplasm. 

 There is thus no division of the protoplasm corresponding to 

 that of the two nuclei. In the next figure (17**) the vacuole 

 is bounded at both ends by a mass of finely granular plasm, 

 on the left by the sac-wall, and on tlie right by a delicate 

 layer of plasm connecting the apical nucleus with the basal 

 one. Fig. IV'' shows the protoplasm of the sac aggregated 

 at the two ends and tlie vacuole in direct contact with the 

 cell-wall. 



What may happen next is the formation of a membrane 

 {m, fig. 17'"), wliich separates the apical part of the ])roto- 



