478 The Ohio Naturalist. [Vol. IX, No. 6, 



The most striking characters noted in the study of the 

 Agrostemma githago referred to above were (a) the formation 

 of an inner and an outer nucellar zone, (b) the growth of the 

 nucellus so as to leave the embryo sac deeply embedded, (c) the 

 formation of a nucellar beak which projects through the micro- 

 pyle, and (d) a filamentous embryo with the large basal cell. 



Since the publication of the above paper, L. S. Gibbs has pub- 

 lished a paper on "Notes on the Development and Structure of 

 the Seed in Alisnoideae"J in which the characters were very 

 similar to A. githago. The development of the embryo sac was 

 the same, also the two zones, which are referred to as nucellar 

 layers, and the beak like projection of the nucellus through the 

 micropyle which is somewhat more pointed than in A. githago 

 and is referred to as a papilla. There is no lateral pouch-like 

 enlargement of the sac as in A. githago. The development of 

 the embryo is practically the same. 



No attempt was made to trace the early history of the embryo 

 sacs in either species. In both species the two nucellar zones 

 and the beak were very evident ; the sacs are located deep in the 

 nucellus but there was no pronovmced lateral enlargement as in 

 the case of A. githago although in some instances there was a 

 slight enlargement of this kind. In both species the embryos 

 were filamentous in their early stages and possess the large 

 basal cells. 



In V. vaccaria the embryo was exceptionally long (Fig. 1), 

 the length being due to the excessive length of the three lower 

 cells. The basal cell becomes very large, stains readily and has 

 very much the appearance of the corresponding cell in A. githago. 

 It is also directly connected with the passage through which the 

 ]x)llcn tube entered and which is now filled with protoplasm. 

 Gibbs says that in Stellaria this basal cell is "elongated so much 

 that it forms a haustorium at the micropolar end, which projects 

 beyond the emljryo sac into the nucellar tissue." Gibbs also 

 says that "the nucellus is very large and active in appearance, 

 and the cell suggests an absorbent organ." 



In Silene conoidea the embryo is not so long as in V. vaccaria 

 Ijut is almost identical in development with A. githago (Fig. 3). 

 However, instead of the cell next to the apical cell being the first 

 to divide, it is usually the second from the apical cell (Fig. 4). 

 The large basal cell eventually disentegrates (Fig. 5) and the 

 further history of the embryo is practically the same as in A. 

 githago. In one instance an abnormal embryo (Fig. (i) was 

 observed. 



The endosperm in both species was non-cellvilar (Fig. .") and 

 identical with the endosperm in A. githago. 



t Annals of I'.ni;iii,v, X'ol. \.\1, pp. 2.")-55, 1907. 



