„,i.cluding a new Arrangement of Phanerogamous Plants. 85 



diacese, the ovules of which have an equivalent character in those 

 genera in which the seed subsequently produced is horizontal ; 

 and even when they are two it is very rare, having been only 

 observed in Talauma among Magnoliacese and Trianthema, the 

 latter of which is afterwards more particularly noticed. 



6. Ovule horizontal with the raphe on the under surface. Of 

 this no instance has been observed where the ovule is single, nor 

 yet where they are two, unless it is that in Geranium Robertia- 

 num they are in their early stages nearly horizontal ; and Ascle- 

 piadeae and Apocynacese are the only ascei'tained instances where 

 the ovules are numerous. 



7. Ovule erect with the raphe next the placenta. This, as is 

 well known, is the ordinary position of the raphe in erect ana- 

 tropal ovules, and occurs I believe not unfrequently where they 

 are numerous, as in Cuphea and Reaumuria. 



8. Ovule erect with the raphe lateral. This character, which 

 was first obsei'ved by Mr. Bennett in Rhamnacete, and by that 

 gentleman attributed to torsion of the funiculus, obtains to a 

 considerable extent among Exogenous families, but is very rare 

 in the Endogenous, Calamus viminalis being the only instance 

 hitherto observed. 



1. Elseagnacese. 2. Rhamnacese. 3. Staphyleaceae (ovules 

 in two rows). 4. Stilbacese. 5. Portulacese (ovules campylo- 

 tropal). 6. Justicia (ovules two, one above the other), 7. La- 

 biatse and Verbenacese ?*. 8. Jasminacese (ovules two). 9. Tri- 

 anthema decandra (ovules campylotropal). 10. Goodeniaceae 

 (ovules two or more, numerous) . 



9. Ovule erect with the raphe turned away from the placenta. 

 Since my first Dissertation on the Position of the Raphe was read 

 at the Linn?ean Society, several additions have been made to the 

 instances of the raphe having this position then particularly ad- 

 verted to, and it is not improbable that others remain as yet un- 

 observed, 1. Limnochaiis Humboldtii (ovules numerous). 2. Pe- 

 ncea fruticulosa (ovules two, PI. II. fig. 8). 3. Geissoloma (ovules 

 four, PI. II, figs. 6 & 7). 4 Berberis vulgaris (ovules two). 

 5. Geranium (ovules two). 6. Nolana. 7. Calytrix virgata 

 (ovules two). 8. Compositse (PI. II. fig. 9). 9. Chrysobalanus 

 (ovules two). 



In Compositse the raphe in several genera examined proved to 

 be always on the anterior side of the ovule, and consequently in 

 relation with the anterior angle of the ovary, and hence it is 

 averse from the placenta, supposing the anterior to be the fertile 



* In such species of Labiatic as I have examined, the raphe is not iu re- 

 lation with the inner angle of the acheniuni, but inclines more or less away 

 from it as it ascends. 



