Mr. C. C. Babington on the British Violets. 101 



gation of minute external differences, the examination of 

 which would probably appear to be little better than a waste 

 of time to a superficial observer; more particularly when, as 

 is often the case, they are so minute as to escape the notice 

 of all except the practised descriptive botanist. It is scarcely 

 necessary to mention instances in proof of so well-known a 

 fact, but still it may be perhaps as well to produce a single 

 example, before entering upon the peculiar subject of this 

 communication. The form and sculpture of the external coat 

 {testa) of the seed is found to distinguish some few orders 

 amongst the Monocotyledons ; it is occasionally of generic 

 value ; in the Chenopodiacece and Polygonece it distinguishes 

 species, but amongst the Caryophyllacece it does not appear to 

 possess sufficient constancy to point out even varieties. 



But to proceed to the subject more particularly before us* 

 In a communication to the Botanical Society at Edinburgh, 

 my friend Mr. Edward Forbes has directed attention to the 

 form of certain curious spurs or appendages attached to the 

 base of two of the stamens and extending into the spur of the 

 corolla, as affording excellent characters for the formation of 

 sections in the genus Viola*, but he has not applied them 

 in the distinction of species. He finds three different forms 

 to pervade all the Violets that have come under his notice, 

 namely, 1st, a rounded spur, such as is found in V, palustris ; 

 2ndly, a lancet-shaped spur, which occurs in V. odorata, V. 

 canina, and V< hirta ; and 3rdly, a filiform spur, as in V. tri- 

 color and V, lutea. In the course of a series of observations 

 which had for their object the application of these characters 

 to the British Violets, I soon found that the direction of the 

 cells of the anthers differed considerably in the several spe- 

 cies, and I have been led to the conclusion that they also may 

 be employed with great advantage in the discrimination of 

 nearly allied species. I have found the cells of the anthers to 

 be always nearly parallel to each other in V. palustris, V, ca- 

 nina, and V. lutea ; and constantly distant below, but con- 

 verging upwards, until their apices nearly, if not quite, touch 

 each other in V. odorata, V. hirta, and V. tricolor. If now 

 we look to the form of the stipules and the presence or ab- 

 sence of an elongated stem, we shall have four sets of cha- 

 racters by which six species may be satisfactorily distin- 

 guished in Britain, nor does it appear to me to be advisable 

 to divide our native Violets into a greater number of species. 



I shall now endeavour to point out the characters of these 

 plants, and hope, by the help of the accompanying illustration, 



* See Ann. Nat. Hist. vol. vii. p. 157. 



