ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 195 



The author is unable to find any reason for this so-called " paired 

 relationship," but believes that it must be regarded as a later develop- 

 ment, not as a primitive condition. 



Embryos of Zingiberaceae and Marantaceae.* — C. L. Gatin con- 

 tributes a note upon the embryos of the ZingiberacesB and the Maran- 

 tacese. The writer finds that in the former the embryo exhibits a long 

 bent axis with the plumule at the bottom of a cavity shut off by a fold 

 from the exterior (as in the Palms) ; the apex is conical and surrounded 

 by the first leaf. The cotyledon has two well-developed vascular bundles, 

 which may unite near the apex ; the radicle is reduced to a central 

 cylinder, but more rarely (as in Hedychium), the lieginning of differen- 

 tiation may be traced. The embryo of the Marantacete is of similar 

 character, but much more advanced, being specially well-developed in 

 Thalia dealbata, where there is a radicle and several lateral rootlets. The 

 formation of the coleorhiza recalls that in the Palms Pinrnv/a and Calamus, 

 but most stronglv resembles that of the (J-ramineae. 



Embryo-sac of Physostegia.t — L. W. Sharp finds that the arche- 

 sporium in Phijsostegia virnmiana, the species examined, consists of a 

 single hypodermal cell, which becomes the megaspore-mother-cell. The 

 latter forms a row of four megaspores, three of which disorganize, while 

 the chalazal one forms the embrvo-sac, which contains an ego;, two 

 synergids, three antipodal cells, which multiply to form several, and two 

 polar nuclei which fuse. During formation of the embryo-sac a lobe 

 develops near its clialazal end, so that the sac consists of two distinct 

 parts joined by a narrower portion. It is probable that double fertiliza- 

 tion of the usual type occurs. The endosperm is cellular from the 

 beginning, the wall accompanying the first division being in the long 

 axis of the sac. The chalazal portion of the sac, or " endosperm lobe," 

 becomes completely filled with endosperm tissue, w'hich invades and 

 destroys nearly all the integument ; the micropylar portion of the sac 

 contains only a few endosperm cells and later disorganizes, becoming 

 obliterated by the encroaching endosperm. The first division of the 

 fertilized egg is transverse, and the chalazal cell, which becomes em- 

 bedded in the endosperm by reason of the elongation of the micropylar 

 cell, develops very regularly into a typically dicotyledonous embryo 

 which displaces nearly all the endosperm. 



Morpholog-y of Seed of Buckwheat. | — Neil E. Stevens, as a result 

 of the study of the development of the seed of Fagoijyrum esciilentum, 

 is convinced that no perisperm is present in this genus. The early 

 development of the embryo is typical in every respect. The endosperm 

 becomes differentiated into two regions, one of which shows no cell- 

 formation ; the outer layer of the cellular portion takes on the appear- 

 ance and function of a cambium-layer which cuts off cells only on the 

 inner side. The outer layer of the nucellus persists in an actively 

 growing condition till the growth of the endosperm is practically 

 complete, and appears to function as a nutritive jacket. In its later 



* Comptes Rendus, cliv. (1912) pp. 35-7. 



t Bot. Gaz., lii. (1911) pp. 218-25 (2 pis.). 



X Bot. Gaz., liii. (1912) pp. 59-66 (8 figs, in text). 



