282 NOTICES OF BOOKS AND MEMOIRS. 



similar morphologically and functionally to that presented by 

 archegoma. The succession of compression being here, as always, 

 towards the locus of fertilization, it follows that any contents 

 remaining in the tube will be squeezed into the embryo-sac. Its 

 ultimate fate is probably absorption. 



The chief exceptions are the following : — When two germs are 

 present in the embiyo-sac,''= as may sometimes be the case, they 

 may both occasionally be fertilized and surround themselves with 

 a cellulose wall, but one always predominates over the other, which 

 finally disappears.! In (rladiolus communis and Crocus vermis the 

 pollen-tube forces its way between the young upper part (filamentary 

 prolongation) of the auxiliaries, and frequently reaches the germ, on 

 the surface of which it expands, or else it remains at a varying height 

 between the auxiliaries. The prolongation remains for some time 

 after the absorption of the latter, and is found clinging to the pollen- 

 tube, which is much swollen after fertilization..! In Ornithogalum. 

 nutans one of the auxiliaries surrounds itself with a cellulose-wall, 

 but never grows further. In (Eiwtliera Drummondii the tip of the 

 pollen-tube is attached solely to the outer wall of the embryo-sac, 

 a method of union which is even more remote in the case of 

 Ricinus, which shows tw^o intervening laj^ers of* cells. ^i 



The author's observations on the formation of adventitious 

 embryos are of the greatest importance, and possess most vital 

 interest. In Funkia ovcita, after fertilization, single cells of the 

 one-layered nucleus near the germ and auxiliaries begin to grow into 

 the embryo-sac, dividing and forming a several-celled knob. In 

 Notlioscordium friu/rans the germ is almost always suppressed in 

 favour of the adventitious growths, while in Citrus it usually 

 grows along with them. The same method of embryo-formation 

 is presented by CaIcho(ji/ne, and in all probabiUty its existence has 

 given rise to the supposition of parthenogenesis in this plant. 

 Parthenogenesis, being a method of growth from the unfertilised 

 germ, is altogether distinct from the process of adventitious forma- 

 tion of the embryo. AVith these observations before us, therefore, 

 it is plain that in future, for the proof of parthenogenesis in any 

 given case, a necessary factor will be disproof of the existence of 



* The raison d'etre of this second germ is easy to understand. Its nucleus 

 is, of course, that one which, under usual circumstances, comes from the upper 

 part of the embryo-sac to unite with one from the lower part into the special 

 sac-nucleus, but which is retained in position by being surrounded with a 

 membrane. 



+ Orchis latifoUa is mentioned fiutlier on in the book as having two 

 embryoes lormed from two germs. 



+ Strasbnrger supposes that this swelling of the wall of the pollen-tube 

 favours the jmssage of its fertilizing matter. We venture to think that a some- 

 what similar instance is furnished by those seeds which swell up in water — 

 moistening being necessary to the passage of gases through tjje testa. (See 

 Deherain and Landzin in Ann. Sc. Nat., 5me Serie, t. 20, p. 307). The most 

 remarkable occurrence in this connection is the differentiation of the inner , 

 layers into a spiral coil, which is liberated as such on applying moisture. 



§ In view of the function of the thickened upper part of the auxiliaries, 

 which is to render pervious the top of the embryo-sac, this would appear to be 

 either the relic of, or a reversion to, au ancestral condition. 



