378 THE PHENOMENON OF 



vesicle remains wholly free during its development into 

 suspensor (vorkeim) and embryo, becoming developed 

 without any connection with the other phenomena of 

 cell-formation in the embryo-sac, even interfering with 

 and displacing and destroying them, so that it affords us, 

 not merely in its first formation, but also in its further 

 behaviour, the example of the freest and most inde- 

 pendent cell-formation which the plant exhibits. 



j3. Formation of transitory cells in the embryo-sac. — 

 In addition to the germinal vesicles not arriving at 

 development, already mentioned, other cells are met with 

 in the embryo-sac which only show themselves for a 

 short time; free cells, which are dissolved again and 

 vanish without leaving a trace when scarcely formed, or 

 even before completion, as mere nuclei. Among these 

 are especially the cells presenting themselves at the lower 

 (chalazal) portion of the embryo-sac, agreeing more or 

 less in number and form with the germinal vesicles, but 

 never developing into embryos, and which are described 

 in Hofmeister's often-cited essay in the majority of the 

 plants investigated by him.* These strange antipodes 

 of the germinal vesicle originate either simultaneously 

 with the latter, or somewhat later,f and, in exceptional 

 cases, even earlier ; | they mostly disappear again before 

 fertilization, § often even as mere nuclei; in rarer cases 

 they persist for some time after the fertilization, and 

 vanish during the formation of the endosperm-cells. || 



y. Formation of the endosperm through free cell- 

 formation in the embryo-sac. — The vitality of the embryo- 

 sac is retained uninjured as a whole, in spite of the 



* They appear to be absent from Polygonum orientate, (t. xii, f. 18, 19), 

 Agrostemma Githago (t. ii, f. 18—22) Godetia, (t. v, f. 4, 5); they are 

 especially numerous iu Sici/os (t. xiii, f. 3 — 5) ; often only a single one, but 

 of unusual size, presents itself in Hi/acinthus orientatis (t. vi, f, 3). 



t L. c, Orctiis, t. ii, f. % 3, 4. 



j L- c, Crocus, t. \Y, f. 14. 



§ L. c, Ecbatimn, t. xiii, f. 6 — 9. 



II L. c., Eyacintlms orientatis, t. vi, f. 4, i; Hetiatithiis annms, t. xiii, 

 f. 18—20. 



