86 THE JOURNAL OF BOTANY 



mosses and other plants. Under these conditions the variety 

 occurs as a moderately tall, thickly tufted plant, bearing a large 

 number of flowers. In less favourable slacks, where the degree 

 of moisture is slight and fluctuating, and the layer of decayed 

 vegetation is consequently thin and scanty, such plants of 

 Pamassia as occur are squat in habit, with fewer flowering 

 stems, but the flowers are often remarkably large. The older 

 slacks containing dense herbage, and especially those yielding 

 thickets of Salix repens, supply other ecological conditions w r hich 

 modify the habit of the Pamassia. The plants here become 

 " drawn up," being usually taller, having longer leaf -petioles, 

 and the bract placed higher on the scapes. As the " drawing 

 up " effect, of course, acts uniformly on both scapes and petioles, 

 the leaves and bract maintain their relative positions, and do 

 not become separated by a considerable interval, such as usually 

 occurs in elongate specimens of the typical plant of the uplands. 

 The variety is therefore still readily distinguishable, though its 

 normal facies is slightly altered. 



The size of the flowers, again, a feature wmich strikes the 

 casual observer at once, has been remarked to be inconstant. 

 The flowers do, as a matter of fact, vary a good deal in size, 

 both under normal and abnormal conditions. Normally, the first 

 flowers to open are always large, and they may continue to be so 

 until a dozen or more have opened ; by this time the earliest ones 

 will have been fertilized, and the fruit will have commenced to 

 develop. As the fruit ripens and becomes elevated above the 

 younger flowers, the latter begin to show signs of diminution in 

 size, accompanied by a reduction in the thickness of the scapes ; 

 and the plant ultimately becomes so exhausted that the latest 

 blooms may be as small as in typical P. palustris. A similar de- 

 crease in the size of the flowers occurs naturally in many other 

 plants in which the flowering period is protracted, and continues 

 simultaneously with the formation of fruit. It is well exemplified 

 in several varieties or subspecies of Pansies. 



More abnormally, the flowers may be reduced in size by 

 starvation, due to lack of nourishment. Sometimes the seeds 

 germinate where the soil is little but loose sand. Or the plant 

 may commence luxuriant growth in an inundated slack, and 

 throw up many steins. A period of drought supervening while 

 the flower-buds are still very young may dry up the slack, and 

 leave the plant with an inadequate supply of moisture. The 

 more stems the plant has, the less now is the prospect of 

 its flowers attaining their normal size and beauty. We have 

 seen plants when left thus in an exceptionally precarious 

 situation with flowers about half the usual size, and of a 

 sickly greenish-white colour, very different in appearance from 

 the usual handsome w 7 hite blossoms. Even then, the stout 

 rigid stems, much divided crowns, and squat foliage, w r ould be 

 sufficient to indicate the sand-dune variety to anyone familiar 

 with the plant. 



That the var. condensata should so well maintain its indi- 



