Miyake, The development of the gametophytes etc. 13 



Enveloping the archegonial complex, there is a singie layer 

 of sheath- or jacket-cells. At first they are poor in contents and 

 can scarcely be distinguished from the neighboring cells of the 

 prothallium, but later on the cells become rieh in cytoplasm and 

 the miclei more piominent (figs. 69. 70, 72, 87). In the full-grown 

 archegonia, many of the jacket-cells are binucleate. The same 

 was found to be the case with Lihocedms by Lawson (1907). 

 Coker (1903) also nientions that the jacket-cells of the mature 

 archegonia in Tcuodiuni generally contain two nuclei. In Cryp- 

 tomeria, according to Lawson (1904b), nearly all of them are 

 multinucleate. It is to be noticed that the jacket-cells near the 

 apex of the archegonia are generally poor in contents and resembles 

 closely the adjacent prothallial cells. Judging from the figures of 

 Lawson (1904b, figs. 39, 40) this seems to be true also in Cryp- 

 tomeria. 



As the central cell of the archegonium reaches its füll size, 

 the cytoplasm becomes densely granulär and most of the smaller 

 vacuoles disappear, leaving usually one big vacuole at the center. 

 The nucleus now undergoes division. The early stages of the 

 division were not found. Various stages of the karyokinetic spindle 

 are shown in figs. 77 — 81. All of the nuclei of a Single arche- 

 gonial complex seem to divide almost simultaneously. Of the two 

 nuclei thus formed, the Upper one, the ventral canal-uucleus, 

 usually soon degenerates and its remnant may, for a time, be seen 

 as deeply staining body at the tip of the Q^g (fig. 82). 



Arnoldi (1900) denies the forniation of a ventral canal- 

 nucleus in Sequoia, Taxodiuftn, Cryptomeria and Cunninghamia. 

 His conclusion was not confirmed by the later researches of Coker 

 (1903) in Taxodium and those of Lawson (1904) in Sequoia and 

 Cryptomeria. Coker has studied carefully the division in the 

 ventral canal-nucleus in Taxodiuni. Although Lawson did not 

 observe any division-figure in the two genera above mentioned, he 

 seems to have enough evidence for the existence of such division. 

 My observations now put the formation of the ventral canal-nucleus 

 in Cunninghamia beyond doubt. A doubt was also expressed by 

 Arnoldi (1899b) as to the presence of a ventral canal-nucleus in 

 the Cupressitieae; but the recent Investigators (Land, 1902; Noren, 

 1907; Lawson, 1907) all agree in thß existence of such nucleus 

 in that group. Only other species of the Coniferae, in which the 

 absence of a ventral canal-nucleus was reported, is Torreya taxi- 

 folia (Coulter and Land, 1905). It was, however, found in 

 Torreya californica by Miss Robertson (1904), and it is not im- 

 probable that the later researches may reveal the existence of the 

 nucleus in the former species. It seems, therefore, that the for- 

 mation of the ventral caual-cell or nucleus is a rule among the 

 Coniferae and also among the rest of the Gymnosperms. 



The lower of the two nuclei, resulted from the division of 

 the central nucleus, the egg-nucleus, immediately begins to enlarge, 

 and at the same time moves downward (figs. 82, 83). In the 

 mature Qf^g^ the nucleus is usually found about one third below 



