The Morphology of Riippia Maritima. 137 



and I have, therefore, no occasion to alter or add to Murbeck's 

 excellent description. 



In brief, the changes are very similar to those which lead up to 

 the first division in the pollen mother-cells. The megaspore mother- 

 cell and its nucleus enlarge, while the latter goes through the 

 synapsis and succeeding stages, the staining reactions being 

 essentially the same and even the fine kinoplasmic fibers appearing 

 in the cj^toplasm the same as in the corresponding stages in the 

 pollen mother-cell. 



In the spindle formed for the first reduction division, eight chromo- 

 somes appear (PL XII, fig. 80), as Murbeck also announces in 

 Ruppia rostellata. Although the chromosomes are here much thicker 

 than in the sporophytic karyokinesis, they are nevertheless still so 

 small that any definite declaration concerning their shapes and 

 method of splitting is well nigh impossible. Still, as Murbeck has 

 noted, the ring and Y-shaped forms characteristic of the heterotypic 

 division are occasionally apparent. 



After the nucleus of the megaspore mother-cell has divided, we 

 find a wall laid down separating the two daughter-cells (cf. Mur- 

 beck, 1902, fig. 45). In this connection, reference might be made 

 to the case already noted under the megasporangium f'p. 136), where 

 a double megaspore mother-cell was found, producing in each case 

 two such daughter-cells (PL XL fig. 78). 



The second division follows closely on the first, with a ver}- slight 

 pause, similarly as in the pollen mother-cells. The two walls re- 

 sulting from these divisions are, however, laid down at quite different 

 planes with respect to each other, that is, the wall dividing the 

 two inner cells is periclinal, while that separating the two outer is 

 anticlinal. Thus the two outer cells are both in contact with the 

 third cell, and are separated by it from the innermost ceU. The 

 plane of the anticlinal wall is, however, obliquel}- situated with 

 regard to the plane of a horizontal or vertical median section of 

 the megasporangium ; in other words, it is oblique to the plane of 

 the paper on which such a section is represented, so that an ob- 

 lique position of the two upper cells with respect to this plane 

 results. 



This arrangement is shown by Murbeck (1902, pp. 13, 14, 

 fig. 51). But often, due partly to the manner of cutting the section 

 and partly to the orientation of the outer anticlinal wall, the pos- 

 ition and even the number of cells is not so apparent, since one 

 of the two upper cells then lies more or less completely over the 

 other. Such a case is represented in PL XI, fig. 79, which a hasty 



