The Scottish Naturalist. 159 



It is hardly different when we examine the Ordinal mark itself. 

 There are here both superabundance and deficiency. Serial and 

 Cohortal characters are reproduced as Ordinal ; we are treated to 

 the contradictory alternative ; and various characters (such as 

 " anthers innate ") will not stand the application of the quantita- 

 tive test. 



A similar criticism holds when we advert to the genus Ranun- 

 culus. Much that is properly Tribal, and much that is Cohortal, 

 are here mixed up with what is strictly generic ; while several 

 Generic characters are conspicuous by their absence. Further, we 

 have more than one striking instance of the absurdity of contra- 

 dictory opposites. 



But, lest I should appear to exaggerate, it may be well to give 

 Hooker and Bentham's generic mark in extenso. Here it is, with 

 the added strictures enclosed in square brackets : — " Sepals 3-5, 

 deciduous [this last is ordinal] \ petals as many or more (up to 15), 

 with basal gland or fovea, with or without a scale [contradictory 

 alternative, hence no character at all], conspicuous or more rarely 

 minute ; carpels indefinite, uniovulate [tribal] ; ovule ascending 

 from base of cavity [tribal] ; achenes capitate or spicate, apiculate 

 with short or very short style, or beaked with longer. Herbs 

 annual, or very often possessing a perennial stem [ordinal] ; leaves 

 entire or cut [contradictory alternatives], cauline very often with 

 few divisions ; flowers [should be petals] white [only in limb], 

 yellow or red [flowers], terminal, solitary or paniculate, more rarely 

 sessile in the axils of the twigs ; stamens shorter than the sepals or 

 the petals, very frequently numerous [ordinal], sometimes in small- 

 flowered species very few in number ; ache?ies compressed or sub- 

 globose, smooth or variously striated, ribbed, wrinkled or prickly." 



It needs only a glance at this list to see how mixed and unsatis- 

 factory it is, and how multifariously it transgresses the laws of 

 methodic procedure. Omissions also may be detected in it ; e.g., 

 no notice is taken of the hollow stem : and we are not told on 

 which surface of the petal the gland is found. 



One strong feature of the " Genera Plantarum" however (from 

 the side of Method of course), is the scrupulous and consistent 

 noting of exceptions. But surely the plan would be more effective 

 if these exceptions, instead of being relegated as " Abnormal 

 Forms" to a concluding paragraph, and printed in small type, 

 were inserted in the Definition proper. Thus only, as we have 

 already seen, can degree of generality be indicated in a way that 

 shall be clear and satisfactory. 



What then is the upshot of the whole matter? It is simply this. 

 Three things are indispensable in order to the satisfactory handling 

 of the Definition in Botany or in Zoology — viz., to make the list 

 of characters as complete as may be; to indicate degrees of 

 generality by the marking of exceptions on the plan already indi- 

 cated ; and to avoid mutually contradictory alternatives as char- 



