830 SUMMARY OP CURRENT RESEARCHES RELATING TO 



tion of the endosperm, perishing after this is completed. In Liliacofe, 

 JuncagineaB, Iridese, Amaryllidege, and doubtfully in Cjcadese, the 

 cotyledon remains anatomically unchanged during germination, 

 absorbing the endosperm through the ordinary epidermal cells which 

 differ in no way from those of the young leaves. 



In GraminesB, Palmae, Commelynacese, Cyperacege, and Juncaceee, the 

 cotyledon developes special organs or haustoria for the absorption of 

 the endosperm. In grasses this is the scutellum, a shield-like organ, 

 clothed with an epithelium composed of elongated thin-walled absorp- 

 tive cells at right angles to the surface, from four to ten times as long as 

 broad, and projecting into the endosperm like sacs. In Palms and 

 Commelynaceee, on the other hand, the absorptive organ is of the same 

 shape as the seed, its periphery consisting of elongated thin-walled 

 cells placed vertically to the surface, and from two to six times as long 

 as broad. In Cyperacese and Luzula ( Juncacese) the organ in question 

 is of a filiform-cylindrical shape, continually renewing itself at the 

 apex. The entire haustorium consists of elongated thin-walled cells, 

 from four to six times as long as broad, and having their longer. axis 

 parallel to that of the organ. In Juncus the haustorium is pear- 

 shaped ; both the inner and the epidermal cells are elongated in a 

 direction parallel to its longitudinal axis ; the terminal cells at its 

 apex being elongated radially and club-shaped. 



A second type occurs in Dicotyledons and Coniferse. The coty- 

 ledons remain only for a time in the seed, consuming the endosperm, 

 after which they rupture the testa, emerge above the soil, and perform 

 the function of assimilating organs, with the characters of ordinary 

 leaves. Their epidermis consists of thin-walled cells, not specially 

 elongated, but with the ordinary forms of young epidermal cells. 



The main point of difference is that in Monocotyledons the coty- 

 ledon servos only for the absorption of the endosperm, and may deve- 

 lope into a special haustorium ; while in Dicotyledons and Conifers 

 the cotyledons serve in the first place for the absorption of the endo- 

 sperm and afterwards for assimilation, when they assume the form 

 of ordinary leaves. 



Embryo-sac of Santalum and Daphne.*— Prof. E. Strasburger 

 has made further observations on the structure of the embryo-sac of 

 Santalum album from Madras. In contrast to previous statements, he 

 finds that the " egg-apparatus " follows the ordinary rule of possessing 

 only a single ovum-cell and two synergidse. A fallacious appearance 

 is presented by the synergid-caps (or filiform apparatus) being 

 strongly separated from the synergidee themselves, and a ridge 

 springing out from the wall of the embryo-sac between them. The 

 pollen-tube forces its way between the two synergid-caps to the point 

 of insertion of the ovum-cell, after which the egg-apparatus exhibits 

 its ordinary changes. The formation of endosperm is commenced by 

 the division of the secondary nucleus of the embryo-sac, and a 

 transverse division of the embryo-sac itself beneath the swollen spot. 



With regard to the embryo-sac of Daphne, Strasburger admits the 



* Ber. Deutsch. Bot. GeselL, iii. (1885) pp. 105-13 (1 pi.). 



