Mr. Bentham's Rn>iew of the Order o/" Hydrophyllese. 271 



tinue to line the capsule without adhering to it, forming, as it were, an inner 

 capsule, and at complete maturity diy up into a thin membrane. In Eutoca, 

 Phacelia, and Emmenanthe they are converted into spurious dissepiments, 

 wiiich in some species meet in the centre so as to divide the capsule into two 

 cells, and in Emmenanthe are, moreover, considerably dilated in the centre. 

 In Eutoca Mexicana the adhesion with the parietes breaks off, and the fruit 

 assumes the appearance of an unilocular polyspermous capsule with two cen- 

 tral placentae. The same thing appears to take place in Phaceliajimbriata, 

 but I have not seen any capsule in a state far enough advanced to be certain 

 of the fact. 



The form of the capsule, ovoid or globose in most Hydrophyllece, is oblong- 

 linear and compressed in Emmenanthe. 



The number of ovulae appears to have been the character chiefly relied upon 

 in the formation of the genus Eutoca, and is, indeed, the only one which 

 separates it from Phacelia, there being two only to each placenta in Phacelia, 

 and often a great number) in Eutoca. This character, however, is very uncer- 

 tain, and forms very unnatural groups, whether we draw the line at 2, 4. 6, or 

 8 to each placenta, or between the definite number, arranged in two rows on 

 the one side, and the indefinite number, without apparent arrangement, on the 

 other. The same character also, if applied to Nemophila, would dissever N. in- 

 signis from A^. phacelioides, and A'^. pedunculata from A^. parvijlora. 



The seeds of all the Hydrophyllece I have been able to examine appear to be 

 the same as those of Eutoca described by Mr. Brown in the above-quoted 

 Appendix to Franklin's Voyage. 



From the above observations, and the characters of the six genera of which 

 Hydrophyllece are now composed, it would appear that Hydrophyllum is a very 

 natural genus, though difficult to characterize. Nemophila and Ellisia, when 

 taken together, are a natural group, but are separated by a purely artificial 

 character, and the same thing may be said of Eutoca and Phacelia. Emme- 

 nanthe consists of but one species, so peculiar in its appearance and several 

 characters that it will probably always remain distinct and well marked. 



I now proceed to a short synopsis of the genera and species of which the 

 order consists. 



