310 ON THK GERMINATION OF CHAEA, 



indicated, without any fixed rule as to position. jN'ordstedt states 

 that among 130 germinating plants he observed one instance in which 

 there were five accessory or adventitious pro-embryos, three in which 

 there were four, three in which there were three, nine in which there 

 were two, and twelve in which there was one accessory pro-embryo 

 from the base of the primary root. Among five young germinating 

 plants of Chara crinita (which is represented in an advanced stage 

 in fig. 6), two were observed to be destitute of an accessory pro-embryo, 

 two with a single one each at the base of the primary root, and one 

 with one here and three on the second node. The most prolific among 

 those examined were the Tolypellas ; and one plant of T. glomerata, 

 which I cite as an illustration, had five accessory pro-embryos from 

 the base of the primary root, and three from the second node. It is 

 scarcely necessary to say that the accessory pro-embryos are capable 

 of development in the same manner from the nodes of their roots. 

 This was actually seen in the plant of Tolyp. gloinerata, to which 

 reference has just been made, in -which two of the accessory pro- 

 embryos issuing from the base of the primary root had each given 

 birth to two each, and the other to one, making a total of fourteen 

 pro-embryos on the same plant, if we include the primary or principal 

 one. Nordstedt asserts that it is not uncommon to find a whole dozen 

 accessory pro-embryos on the first node of Tolypella nidifica, A. Br., 

 besides accessory branches and ;roots. In this manner the plant soon 

 forms numerous stems. 



It now remains for us to say something respecting the relative 

 position of the successive divisions, and the direction of growth of the 

 various parts originating in the plants under observation, in order to 

 complete the preceding representation of the building up of the 

 Characem. 



The first partition which cuts off the primary node stands almost 

 or quite perpendicular to the greater axis of the oospore, and the 

 second, or halving- wall, is always at right angles to the first. Of 

 course there is also a constancy in the position of these walls in rela- 

 tion to the axis of origin, that is, the leaf which bore the oospore ; 

 for the first, an invariable one in every respect, though not for the 

 second, because it may, under conditions already def-^ned, present 

 every conceivable inclination to the plane of the radial 

 longitudinal section of the leaf. An observation of Nordstedt's 

 appears to give the clue to a fixed rule in a certain case, that the 

 partition in question always falls within the plane of a radial longi- 

 tudinal section of the leaf to which it owes its origin. He discovered, 

 for instance, that the oospores he examined of Nitella flexilis {N. 

 furciilata, llebb.) had a lenticular flattened shape, and were so placed 

 that the broad face was in the plane of a radial section, and con- 

 sequently the blunt edges directed from and towards the leaf; and the 

 halving-wall always parallel to the broad side, hence also in the plane 

 of a radial section. This same flattening of the oospore is observable 

 in the species of Nitella qM\q(\. io flex His. In N. tenuissima, of which 

 I had a large number of oospores in a suitable condition for exami- 

 nation, I found them invariably flattened, and the majority of them 

 occupying the same position with regard to the axis of their origin as 

 that indicated by J^ordstedt, whilst a smaller number were the 



