EPIPACTIS VIEIDIPLOEA 37 



tliereforc, that these small creatures may carry a few pollen-grains 

 from one flower to another, but whether they are likely to convey 

 them from one ])lant to another is perhaps doubtful, as it is scarcely 

 probable that they would leave a phmt where they have abundant 

 food. If cross-fertilization occurs in this way it is probably acci- 

 dental, self-fertilization is the rule. I repeatedly saw ^. latifolia 

 and E. violacea visited by Avasps, and several times caught these 

 insects with poUinia on their heads. I saw no insects visit viridijlora^ 

 but I had not so many opportunities for watching the latter plant. 



Miiller studied most minutely the reproductive organs of E. viri- 

 di flora, both from a morpliological and physiological point of view, 

 in comparison with E. latifolia. His opinion therefore is of more 

 weight than that of those botanists who have confined their observa- 

 tions to the more obvious characters of the leaves, bracts, and 

 perianth : he says (/. c.) that we have in E. latifolia and E. viridi- 

 flora two form-cycles which diifer from one another by thoroughly 

 essential characters, and have the same claim to be considered distinct 

 species, as any two species of a genus. 



Explanation of Plate 553. 



A. Ei)ipactisvlridi flora \a,v.leptochila. B. E. latifolia. C. E. violacea. 1. Side 

 view of flower, sepals and petals removed. 2. Back view of column. 

 3. Front view of do. : a., anther ; p., pollinia ; o., ovary ; r., rostellum ; 

 s., stigma ; st., staminode. 



THE UREDINEyE OF WEST SOMERSET. 

 By NoRMAif Gr. Hadden. 



As there appears to be no published account of the Rust Fungi 

 of the district with which this paper deals, it may be as well to put 

 on record those species which I have observed here during the last 

 four years. The district covered is a narrow strip about twelve miles 

 in length and four in width, forming the extreme north-western 

 corner of Somerset : it is bounded on the north by the Bristol Channel, 

 on the south by Exmoor Forest and on the west by the Devon county 

 boundary-line. Taking the village of Porlock as the centre, all the 

 species recorded have been found within a few miles' radius. Owing 

 to the variety of plant associations in the neighbourhood, the list of 

 TTredinece is a long one and includes a number of rare and interesting 

 species. The salt-marshes, rich pasture land, large old woods, young 

 plantations and open moorland naturally support a great variety of 

 phanerogamic plants with the consequent number of parasitic rusts. 



I am greatly indebted to Mr. Carleton Rea for his kindness in 

 assisting me in the determination of some of the more critical species. 



The nomenclature adopted in the following list is that employed 

 by Mr. J. Ramsbottom in his list of British Uredinales, published in 

 the Transactions of the British 3Iycological Society, vol. iv. : — 



JJromyces FicaricB Lev. On Ranunculus Ficaria, frequent. — • 

 TT. caryophyllinus Scliroet. On border Carnations in my garden— 



