86 The Ohio Naturalist. [Vol. IV, No. 4, 



passes in its development from orthotropous, through the anatro- 

 pous condition and becomes campylotropous when mature. The 

 setting aside of the lower endosperm nucleus of the first division 

 b}' a wall and its development as a large vesicular cell, as is the ^ 

 case in Sagittaria and Vallisneria, while not confined to the Helo- ' 

 biae nor apparently characteristic of all of them, is yet significant 

 when one finds a similar peculiarity in some of the Nymphaeaceae. 



The number of ovules in the carpel also shows a diverse devel- 

 opment. In Potamogeton, Alisma, Sagittaria and Nelumbo there 

 is usually a single ovule in each ovulary. In Butomus, Vallisneria 

 and other genera of the Hydrocharitales the ovules are scattered 

 on the inner surface of the ovularies. This is also one of the 

 striking characters of Nymphaea and Castalia. In fact the pecu- 

 liar way in wdiich the ovularies of certain Nymphaeaceae agree 

 with many of the Hydrocharitales must appear most interesting 

 to anyone who has made the comparison. Coalescence and 

 epigyny also figure in both the Hydrocharitales and Nymphaea- 

 ceae. Thus it will appear that what might be considered as minor 

 or secondary characters do not detract but rather add considerably 

 to the weight of the argument that the Nymphaeaceae have very 

 much in common with the Helobiae. 



The important investigation of Lyon shows conclusively that 

 he was correct in claiming that the embryo of Nelumbo is essen- 

 tially of the " Monocotyl" type, and, since there can be no ques- 

 tion of the facts, it also appears that his further conclu.sion was 

 unavoidable that the Nymphaeaceae should be placed near the 

 Helobiae. As stated before, the writer had the opportunity of 

 studying .some of Cook's preparations and it became evident that 

 the embryo of Nymphaea advena is in all essentials similar to that 

 of Nelumbo. Later a special study was made of the young embryo 

 of Nymphaea advena. As stated by Cook, in the young embryo 

 of Nymphaea the "cotyledon" is at first not lobed. Later there 

 is a rapid development at the two sides resulting in a two-lobed 

 structure (Figs. 14, 15). Since Conrad stated that in Castalia 

 odorata the embryo has two distinct "cotyledons" from the first, 

 a study was also made of this species. With some difficulty very 

 young embryos were dissected out of their embryosacs which are 

 easily removed from the ovule. It was found that although the 

 "Dicotyl" a])pearance is quite strong, the embryo must be 

 regarded as of the same type as Nymphaea and Nelumbo. In 

 the very young embryo there is an expansion which extends 

 nearly around the base but is discontiiuious at one side ( Fig._ 16). 

 vSoon this expansion develops on oi)posite sides as two pronnnent 

 lobes in such a manner that the original connection between the 

 two lobes is very difficult to distinguish (Fig. 17). On examining 

 the embryo from below, however, the similarity to the Nelumbo 

 and Nymphaea embryos becomes perfectl)' apparent (Figs. 18, 



