266 The Irish Naturalist. [Oct., 1S97. 



counted at least eight patches of the plant, some of them very 

 large, the largest covering an area of upwards of twelve square 

 yards. In three of these plots the general aspect of the plant 

 differs from that of the others in that the stems are less decumbent, 

 the form of leaf narrower and more linear, and the internodes 

 of the primarj^ flowering stems longer. In all, there is some 

 dissimilarity in the leaves, not only in the same stem, but to 

 some extent even in the same whorl. Onlj' in the lower 

 whorls of some stems has the form of leaf any appreciable 

 tendency towards being obovate, a tendency which may be 

 noticed also in those of the lowest whorl of other species of 

 the same genus, but these leaves are evanescent. The 

 variation in form of leaf in plants in different parts of the 

 field, ranging from linear oblong to comparatively broad 

 lanceolate, cannot be held to constitute any specific or even 

 varietal distinction, and is probably to be attributed to the 

 nature of the soil and other local conditions. In all, the 

 fruiting branches of the panicle are erect, or sub-erect, with 

 trichotomously branched umbels. The leaves of all in the 

 earl}' stage of growth are erect, after^vards spreading or erecto- 

 patent, and it has been noticed that in some of the main 

 flowering stems, especially in those of the less slender forms, 

 the}^ become reflexed. In the plant which I regard as normal 

 G. ereduvi, as also in the more robust forms, the corolla is pure 

 white and the leaves grass-green ; in some others the flower 

 is slightly cream}^ and the foliage a yellowish green. The 

 short growth, which was at first supposed to be from the base 

 of the main stems, giving a tufted and matted appearance to 

 the patch, is for the most part an independent growth of 

 short, wxak, sterile stems, many of which fade about the time 

 the flower-buds of the primary stems begin to expand, which, 

 in the case of the Glenmore plant, did not occur this 3'ear 

 until late in July, though at Whitewell and Cullybackey it 

 w^as observed in flower in the latter part of June. The 

 inflorescence has a pleasant tragrance, resembling that of 

 dried Woodruff". The number of leaves in a whorl being very 

 variable, is not to be depended upon as a distinctive mark of 

 any importance. In the Glenmore plant there are as few as 

 five, and as many as nine, and in the Whitewell plant there 

 are many whorls with ten, and some with twelve leaves, but 



