ADDITIONAL SPECIES, ETC. 321 



a species which had heen separated from the original V. 

 caiiina so long ago as the time of Dillenius, and had been 

 specially named " flavicornis " by Smith more than a 

 quarter century ago. By whatever name to be designated, 

 however, the distribution of the i^resent species stni re- 

 quires to be more satisfactorily determined ; and therefore 

 in reporting localities for it, botanists should take care to 

 let us know the species really intended, which the name 

 "canina" cannot do without explanation, as it may now 

 signify either sylvatica or flavicornis, or even lactea. Mr. 

 Babington also suggests that V. montana (Linn.) may be 

 a variety of his V. canina (Man. ed. 3), but the V. montana 

 of the Linnean herbarium appears widely different. (See 

 vol. i. page 177.) 



135*. Viola lactea, " Smr—Borr. 



Area 12*4*****^*** [14]. 



South limit in Devon, Sussex. 



North limit in Norfolk, ? 



Estimate of provinces — ? Estimate of counties — ? 



Latitude 50 — 53. English tj^pe of distribution. 



Agrarian region. Inferagrarian zone. 



Descends to the coast level, in Ouse. 



Ascends to 50 yards, less or more. 



Range of mean annual temperature 51 — 49. 



Native. Uliginal or Ericetal. I am induced to treat 

 this apart from V. flavicornis and V. stagnina, although 

 uncertain whether all the plants thus named might not 

 more correctly be divided between those two sjiecies. It 

 is highly probable that Smith himself never understood 

 his own V. lactea ; and certainly most of the stations on 



VOL. III. 2 T 



