Aug.. 1893- MONOCOTYLEDONOUS PLANTS. 131 



which are characteristic of the adaptations of exogens to an aquatic, 

 and, in some cases, to other habits of Hfe." 



As regards the points of difference between the two groups, a 

 most important one hes in the structure of the embryo, in the presence 

 of one or two cotyledons. To quote again, " assuming a mono- 

 cotyledon to have descended from a plant with two cotyledons through 

 an aquatic habit, do we find any instances in support of this view 

 among existing dicotyledons, which are now aquatic, or have 

 presumably descended from water-plants ? " Tnipa natans is instanced. 

 The view generally taken of the embryo of this plant is that there is 

 one perfect cotyledon, and one rudimentary. Drs. Gibelli and Ferrero, 

 who have lately studied the embryology, consider, on the other hand, 

 that the mature embryo is an amorphous mass or true thallome on 

 which is developed a single bud, the plumule ; it is, in fact, a degra- 

 dation resembling that of parasitic and semiparasitic plants, such as 

 Orchids, Orobanche, Balanophora, etc., and of some aquatic genera, 

 e.g., Zostera, Hydrocharis and Utricularia. Hence, from the example of 

 Trapa natans it is quite as legitimate, perhaps more so, to conclude 

 that endogens originated from exogens through a parasitic as through 

 an aquatic habit. Ranunaihis Ficavia is given as another case. It 

 is admitted, however, that it is not a true aquatic plant, but has, the 

 author thinks, without doubt descended from an aquatic ancestor, 

 one of the reasons being that it has only one cotyledon ! This may 

 be, but it is rather begging the question to use it as an example in 

 support of the aquatic theory. 



The most instructive instance, however, it is said, would seem to 

 be that of Carnm Bulbocastanuni, which has a monocotyledonous 

 embryo, and an embryology almost identical with that of Sparganiuvi 

 ramosum, while its final form is that of the wheat embryo. Cariim 

 Bulbocastanuni, however, inhabits chalky fields, and, like C. Cariii, in 

 which also one cotyledon is frequently rudimentary, is decidedly not 

 aquatic. Again, the difficulty is got over by supposing that these two 

 species, as well as other umbelliferous genera with linear cotyledons 

 and finely dissected foliage, were ancestrally aquatic ; but linear 

 cotyledons are no proof of an aquatic habit, as will be seen by turning 

 over the pages of Sir John Lubbock's " Seedlings " ; in the case of 

 umbellifers the shape may facilitate their escape from the often tough- 

 walled fruit. Hydvocotyle vulgaris, the aquatic umbellifer pay excel- 

 lence, has ovate cotyledons. Anyhow, we can hardly accept these as 

 evidence. 



Utyicularia, again, has the undifferentiated embryo so charac- 

 teristic of Orchids and other parasites, and cannot be classed as 

 monocotyledonous. Erigenia bulbosa, an anomalous umbellifer, growing 

 in damp and flooded places, has also a very reduced embryo. 



In Cyclamen the cotyledons first appear as two small eminences, 

 one of which grows no further, while the other forms a large green 

 leaf, resembling in all respects a foUage leaf. Gressner, however, 



K 2 



