630 MOSSES AND FERNS 



P. 276. The prothallium of Angiopteris (see Campbell (33) ) 

 not infrequently has the usual heart-shape, or may even be consider- 

 ably elongated. Where fertilization is prevented, it may reach a 

 very large size. Gametophytes of three centimeters or more in 

 length have been observed by the writer in Danaa, and almost as large 

 ones in Kaulfussia and Macroglossum. These large prothallia are 

 often branched, four growing points being noted in one case. (For 

 details see Campbell (33).) 



P. 280. The archegonium of the other genera closely resembles 

 that of Marattia. In Kaulfussia it is rather larger, and in Danaa 

 the ventral canal-cell is very diflScult to demonstrate, indeed, it looks 

 as if it were absent in many cases. In this respect, Danaa recalls the 

 behaviour of Ophioglossum. 



P. 281. The writer has investigated the development of the -embryo 

 in all of the genera except Archangiopteris. (See Campbell (33, 36).) 

 There are some marked differences shown in the different genera. In 

 all cases the primary (basal) wall is transverse, and in Marattia, 

 Kaulfussia, and Angiopteris the whole of the egg takes part in the 

 development of the embryo ; but in Dancea and Macroglossum there 

 is a suspensor formed. In the former the fertilised egg elongates be- 

 fore the basal wall is formed, and the cell next the opening of the 

 archegonium, i.e., the lower or hypobasal cell, develops into a short 

 suspensor, while the whole of the embryo proper is derived from the 

 epibasal portion of the two-celled embryo. 



In Macroglossum (Campbell (36) ) the suspensor is much larger, 

 but its origin is not quite clear. 



P. 282. In Marattia, Angiopteris, and Kaulfussia the basal wall 

 divides the embryo into two nearly equal parts, the hypobasal cell 

 (that next the archegonium neck) giving rise to a large, nearly hemi- 

 spherical foot; from the inner or- epibasal cell the cotyledon is 

 developed, and later the stem-apex. The young embryo is decidedly 

 flattened at first, but later becomes almost globular, and then elon- 

 gated vertically. At this stage the embryo is bi-polar, as it is in 

 Ophioglossum. 



No trace of a root can be recognised until the embryo has reached 

 a considerable size. Then there may be seen near the junction of 

 the foot and cotyledon, near the centre of the embryo, a group of 

 active cells, which it is soon evident constitute the growing point of 

 the primary root, which is thus seen to originate in exactly the same 

 way as it does in Ophioglossum Moluccanum. A single apical cell is 

 present, which is somewhat variable in form. The root finally pushes 



