THE EMBEYO-SAC IN AKOIOSPEItMS. 535 



In PI. XXIV. figs. 4-6, and PI. XXV. fig. 1, the curvature of 

 the ovule and the uprising of the integument are shown ; one cell, 

 filling up the narrow apex of the ovule, enlarges and keeps its 

 position as sole occupier of the fore part of the nucellus (PI. XXIV. 

 fig. 6). Its contents are coarsely granular ; and its large sphe- 

 rical nucleus has several distinct granules inside. In the epider- 

 mal cells one not rarely meets with nuclei presenting the ap- 

 pearance depicted in PL XXIV. fig. 7, of which an enlarged 

 drawing is appended : the oval nucleus is traversed by alternate 

 dark and light (granular and pellucid) bands. The one cell 

 referred to, in the apex of the nucellus, grows out, as in Py- 

 rethrum, into the space left by the closing-up integument, and 

 destroys the epidermis as before. It appears to become trans- 

 formed directly into the embryo-sac; but, as already stated, fur- 

 ther stages are required. 



Embryo-sac Sfc. of Lobelia syphilitica. (Plate XXV.) 



The earlier stages of development of the ovule are so like those 

 of the Composites that we need not dwell upon the description 

 of PI. XXV. fig. 4. In fig. 5 the embryo-sac mother cell has be- 

 come divided at its upper third ; in fig. 6 a second division has 

 followed upon this ; and fig. 7 is a similar stage of a shorter 

 stouter ovule. In the last case the lower cell appears to be en- 

 larging at the expense of the upper ones ; and in fig. 8 its con- 

 tents have become rearranged about two nuclei, between which 

 is a vacuole-like space traversed by a delicate division-plate : if 

 this be compared with Strasburger's earlier drawing (fig. 76, 

 pi. ii. ' Ueber Befruchtung ' Ac.) of Orchis pallens, the correspond- 

 ence of the faint line there passing across the " vacuole " between 

 the tw r o nuclei is evident ; in both cases, I believe, we have a dis- 

 tinct cell -wall, and that the difference between this division and 

 those which cut off the two u cap-cells M is only one of degree. 



Similar cases appear to occur in his later drawings, e. g. in An- 

 thericum (pi. vi. fig. 75, ' Die Angiosp. u. Gymnosp.') and Allium 

 (same plate, fig. 82) possibly*. In the figure we are considering 

 (PI. XXV. fig. 8) the two u cap-cells " are seen to persist as re- 

 fractive streaks on the apex of the embryo-sac, as they are also 

 in PI. XXV. fig. 9, where the two nuclei of the sac have passed 

 to each end, and are separated by a large vacuole as the sac 

 elongates. 



* Cf. also Warming, loc. cit pi. viii. fig. 19, the lowest line across the sac. 



