SCIENCE-GOSSIP. 



"5 



felt of the under-surface of the leaves and by the 

 smaller ellipsoidal fruit. The outline of the leaf 

 is also different, being, as a whole, more obovate, 

 and the sides of the lobes less rounded. It would 

 be very desirable to have more details respecting 

 the size of the flower, the colour of the anther, 

 and the shape, size, and colour of the ripe fruit, in 

 the Arran plant ; with regard to which points Dr. 

 Boswell's and Mr. Duthie's observations in the Bot. 

 Exchange Club Rep., 1872-4, page2i, seematpresent 

 the only available ones. Mr. Duthie reports that 

 the flowers have " rather a pleasant odour, resem- 

 bling that of the mountain ash," and that "the 

 fruit is sweet-tasted." Dr. Boswell adds to this 

 that the fruit is "about the size of that of P. 

 rupicola." Shrubbery and plantation specimens of 

 P. scandica are not, at least by leaf-characters, 

 distinguishable from the wild Arran plant. 



Further observation and cultivation are neces- 

 sary to determine whether P. minima would not 

 be better placed as a variety of P. scandica than 

 as a separate species. 



6. P. fennica, Bab. I feel compelled to accept 

 Dr. Boswell's view expressed in the Bot. Exchange 

 Club Rep., 1872-4, page 22, and to regard the plant 

 of Glen Catacol, Arran, which he describes under 

 this name, as distinct from P. scandica. The 

 surface and clothing of the leaf, as well as its 

 cutting, differs in the two plants. In the present 

 plant the leaf is sub-pinnate at the base, and 

 becomes nearly glabrous with age. The flower and 

 fruit of the two, so far as can be judged from dried 

 specimens with unripe fruit, seem to bear a close 

 resemblance. Dr. Boswell asserts that P. fennica 

 is common as a shrubbery plant. I have never met 

 with it as such, nor seen specimens ; nor do 

 specimens of such shrubbery plants exist in the 

 Boswell Herbarium. The allied shrubbery plants 

 which I have seen are to be referred either 

 to P. scandica or P. pinnatifida. 



7. P. pinnatifida, Ehrh. This is only known in 

 Britain in shrubberies, where, however, it is rare. 

 It is easily distinguished from the last, by its longer 

 leaves with a linear outline ; the base of the leaf 

 pinnate, the long upper part only shallowly 

 pinnatifid. It forms a small tree, with handsome 

 flowers and large red fruit. I have seen an old 

 tree of this form in the garden of the vicarage 

 at Alstonfield, Staffordshire, since unfortunately 

 destroyed by a storm ; small trees in a shrub- 

 bery at Wastdalefoot, Cumberland. 



I wish to acknowledge the great debt which I 

 owe, in writing the above notes, to the work of 

 Mr. N. E. Brown (Eng. Bot., Ed. iii, Supplement), 

 a debt which is none the less real that I am unable 

 always to accept his classification or conclusions. 

 With regard to the plates accompanying his work, 

 I may be permitted to make one or two observations 

 in conclusion. 



Plate 484a (P. intermedia), so far as the leaves are 

 concerned, is a very good representation of the 

 Piercefield Park plant; but the flowers are too 

 small, the filaments of the stamens too short, and 

 the anthers are represented as yellow, instead of 

 pink. The fruit of the Piercefield Park plant is 

 large and round, instead of, as represented in this 

 plate, small and ellipsoidal. Does the fruit here 

 represent that of the Arran P. scandica ? 



Plate 485a (P. pinnatifida). Leaves of the flow- 

 ering stem here represented are just those of the 

 P. scandica of Arran. The free leaf is that of P. 

 fennica, the fruit is a good representation of that of 

 P. minima. I am unable to criticise the flower ; in 

 length and colour of the stamens it is not like that • 

 of P. minima. 



Plate 485b (P. semipinnata) A The leaf and fruit 

 here are a very good representation of P. pinnatifida, 

 Ehrh. I am again unable to criticise the flower. 

 Do any of the British forms exhibit the -yellow 

 anthers represented in all these plates ? 



Sellack Vicarage, Ross, Herefordshire ; June 4th, 1895. 



HABITS OF SPIDERS. 

 By J. Beecham Mayor, L.R.C.P., M.R.C.S. 



"p\R. W. H. DALLINGER has recently written a 

 paper dealing with the constructive ingenuity 

 of spiders, from which we may conclude that 

 they not only inherit, but also acquire, useful 

 and beneficial habits, or modify those inherited. 

 For instance, the triangle spider (Hyptiotes cavatus), 

 of America, so called from constructing a web 

 that is only a segment of a circle, makes the web 

 not only a snare but also a gin, i.e., a stratagem 

 or contrivance closing suddenly upon the snared 

 victim. When, from position or other circum- 

 stances, it has been found necessary to keep 

 the web constantly extended and drawn taut, it 

 is often found that the circular-web spiders will 

 attach a weight to the end of a line connected by 

 cords with the framework of the whole of the web. 

 An instance of this adaptation to environment has 

 recently come under my notice. A lead water- 

 supply pipe runs along the ceiling of a cellar, and 

 from it depended a piece of twisted string, about 

 eight inches in length. The end of this string was 

 drawn upwards by the spider with a strong silken 

 strand attached to the ceiling some little distance 

 away from the leaden pipe, so that the string 

 formed a perfect curve. In the space thus bounded 

 by the ceiling, the string and the silken strand, 

 joining the two latter, a perfect circular web had 

 been constructed by the ingenious spider, the string 

 acting as a weight or counterpoise to keep the web 

 firmly stretched, as well as forming a necessary 

 boundary to it for the attachment of strands. 

 2, St. Alban's ]'il!as. Hcatoii Chapel, Manchester 



F 2 



