No. 454.] STUDIES ON THE PLANT CELL. 741 



no longitudinal splitting of the spirem thread before the second 

 division this mitosis differs from that of the "typical " mitoses of 

 cells and is called "homotypic " to distinguish it on the one hand 

 from the former and on the other from "heterotypic" divisions. 

 Several illustrations of heterotypic and homotypic mitoses to 

 be described presently are presented in Fig. 15, showing the 

 peculiar V-shaped pairs of granddaughter chromosomes, charac- 

 teristic of the first group. It is important to note that whatever 

 the significance of this premature fission of the chromosomes 

 before the second mitosis it is not of the nature of a qualitative 

 reduction division in Weism.an's sense. The details and sig-nifi- 

 cance of reduction phenomena will be considered in other con- 

 nections (Section V). The topics discussed above have been 

 recently studied and reviewed by Mottier (:03). 



We have as yet said nothing of the megaspore mother-cell in 

 Spermatophytes. An increasing number of investigations have 

 clearly established the fact that the embryo-sac in many forms 

 is one of a group of two, three or four cells, each of which is a 

 potential megaspore because its nucleus contains the reduced 

 number of chromosomes. We are accustomed to think of the 

 well known conditions in the lily, where the megaspore mother- 

 cell develops directly into the embryo-sac. But this type with 

 some others {e.g., Fritillaria, Tulipa, Erythronium, etc.) are the 

 exceptions and present a very highly differentiated condition in 

 which the usual developmental history is shortened in a very 

 interesting manner, which will be described presently. 



The embryo-sac arose undoubtedly as one of four megaspores 

 developed after essentially the same manner as microspores or 

 pollen grains, excepting that their arrangement was generally in 

 a row, which is even true of some pollen grains {e. g., Asclepias, 

 Zostera). As stated above, an increasing number of investiga- 

 tions have established the row of four potential megaspores in a 

 large number of forms in various groups. They may not always 

 be distinguished by the form of the group, but their homologies 

 are estabhshed by the mitoses that lead to their differentiation. 

 Two mitoses are of course required to establish the group of 

 four cells and both are identified by the reduced number of 

 chromosomes. Some detailed studies on these mitoses have 



