The Scottish Naturalist. 157 



removed. Moreover, the drawer, when so pulled out, is self- 

 supporting, and does not require to be held up whilst its contents 

 are being examined. This contrivance places within the reach of 

 any visitor collections which would ordinarily require special per- 

 mission to inspect, and that under the supervision of an official. 



The mollusca are as yet only represented by their shells. 

 These are either mounted on cardboard, or, in the case of the 

 smaller species, contained in glass tubes fastened to cardboard by 

 wire. On each card is the name and locality of the species, and 

 the name of the donor. In many cases there are numerous 

 specimens, from various localities, and it is hoped that in course of 

 time the distribution of each species throughout the area will be 

 thoroughly illustrated. Most of the land and fresh water molluscs 

 of Perthshire have already been obtained, and in several cases 

 some interesting varieties or local races are represented also. 



( To be continued. ) 



THE SCIENTIFIC METHOD IN BIOLOGICAL CLASSIFICATION. 



By REV. WILLIAM L. DAVIDSON. 



I. DEFINITION. 



{Continued from page ioi.) 



WELL then, to return to our three denning requisites, I 

 shall try to make .my meaning plain by a few intelligible 

 examples. 



And, first of all, not to go higher than Orders, let us take Ran- 

 unculaceae or the Ranunculus Family. Applying our rules, we 

 find that the mark should run as follows : — " Herbs (except Clema- 

 tideae) with acrid watery juice; leaves radical or alternate (except 

 in Clematideae), exstipulate (except in some of the Thalictra); sepals 

 imbricated (except in Clematideae), deciduous (except in Oxy- 

 graphis, Helleborus, and Paeonia), more than two, petaloid (except 

 in Ranunculeae) ; petals imbricated, (anomalous in Aconitum, Del- 

 phinium, and Aquilegia, wanting in Clematis, Anemone, Traut- 

 vetteria, Caltha, Calthodes, Glaucidium, and Hydrastis) ; stamens 

 numerous (except in Myosurus) ; anthers dehiscing longitudinally ; 

 pistil apocarpous ; ovules anatropal ; carpels free, numerous (except 

 in Actaea), unilocular ; fruit achenes or follicles (coherent in Ni- 

 gella) or berry (in Actaea) ; seeds without an arillus (except occa- 

 sionally in those with follicles) ; testa in monosperms slightly cori- 

 aceous, without prominent raphe, in polysperms crustaceous, with 

 raphe rather prominent ; embryo minute at base of albumen." 



In like manner, if we turn to the tribe Ranunculeae, we shall 

 find the tribal mark, as distin^aished from the others subordinate 

 or superordinate to it, to be: — " Moivers regular; sepals green; 

 petals coloured (rarely, e.g. in Ran. auricomus, unequal) ; carpels 

 developing into achenes, with ovule ascending." 



