512 
On the Pansies of our Coast Sandhills. By J. G. Baxur. 
The fact of the occurrence of two different Pansies, one with 
yellow, and the other with purplish flowers, amongst the sandhills _ 
of the west coast of England, has long been known to botanical 
collectors (vide ‘Cybele,’ vol. i. p. 183), and was two or three 
times mentioned in the old series of the ‘ Phytologist’ (vol. iii. 
p- 46, etc.), but I am not aware that they have ever been fully 
studied and determined. I have never had the opportunity of 
even seeing either of them in a growing state, and can therefore 
only furnish very incomplete information, but possibly a short 
notice here may have the effect of attracting the attention, to the 
matter, of some one more favourably situated for observation. 
It is to the yellow-flowered plant, which grows about the es- 
tuary of the Torridge in Devonshire, and Aberffraw in Anglesea, — 
that the name of Viola Curtisit rightfully belongs. . The other, 
which occurs on the Cheshire coast in the neighbourhood of New 
Brighton, is the V. sabulosa of Boreau (‘ Flore du Centre de la 
France,’ 2nd edit.), and Mackay’s Curtisié from Portmarnock 
appears to be identical. 
The Violas of the Section Melanium of De Candolle may be 
classified under three biological categories. The species of the 
first of these are characterized by their. annual, and of the second 
by their perennial, duration; and they are respectively well re- 
presented in our indigenous Flora by tricolor and lutea. The 
plants of the remaining class are, to quote Jordan, “neither 
strictly annual, biennial, or perennial, but flower usually the first 
year of their existence; and if they live more than a year, their 
root has none the less the aspect of an annual root, and by no 
means -resembles a rhizome which lasts and developes itself each 
year under the ground.” This last mode of growth is exempli- 
fied by V. rothomagensis and several others. To which of these 
categories do Curtisii and sabulosa belong? To the second, I 
suspect ; and if so, geographical considerations afford strong a@ 
priori grounds for a belief that they are not specifically identical 
with lutea. 
As compared with one another, Curtisii more closely resembles 
lutea, and sabulosa tricolor——V. Curtisii: Root long, slender, 
and fibrous. Stems slender, ceespitose. Leaves somewhat fleshy; 
lower ovate, sparimgly crenate; upper narrower. Stipules pin- 
