2 BRITISH EL'i'HKASi^ 



mark. In Wettstein's diagrams of the leaves and bracts this diversity 

 of form is not shown with sufficient clearness, as may be seen by the 

 examination of the actual plants. 



The length of the corolla is variable, and is mireliable as a 

 distinctive character, at least in the British species. For example, 

 E. Kerneri belonging to the Grandiflorce, sometimes occurs with 

 small ilowers, especially at the end of the flowering, and in these the 

 lengthening of the corolla-tube is scarcely apparent. Again, E. lati- 

 foUa normally with the corolla 6 mm. long, has a variety grandiUora 

 'Wetts., w^ith the corolla twice this length. Moreover, the con- 

 spicuousness of the flower is" not dependent on the length of the 

 corolla "" measured along the back," but on the breadth of its lips ; 

 so that a flower may be larger and more conspicuous than its length 

 as given in millimetres would indicate. For this reason and because 

 the measurement of the corolla in the dried state is difficult and 

 uncertain, the flower is, as a rule, in tliis paper, only described as large, 

 small or conspicuous. 



The leaves in simple specimens of any species are usually persistent 

 during the flowering, but in branched j^lants they are deciduous. 

 This is therefore seldom of much value as a specific chamcter. 



The student should be reminded that it is of little use to attempt 

 to name immature, damaged or poor specimens, or those which are 

 past flowering, until he is well acquainted with the species in all its 

 stages. 



After the diagnosis of each species there follow examples of plants 

 from different British localities. These serve as records of stations 

 for ty])ical plants as w^ll as for variations from the type ; in some 

 cases they are of plants which have been distributed, in my opinion, 

 under incorrect names. 



Of my own gathering, all the Scotch and some of the English 

 s])ecimens were named or confirmed by Mr. Townsend, and I have 

 been greatly as.-^isted in my studies by the kindness of Mr. Charles 

 Bayley and Mr. G. Claridge Druce, who have placed their valuable 

 collections of Euphrasia containing many authentically named 

 examples, at my disposal. 



The analytical keys wdiich follow apply only to the normal well- 

 developed form of each species and its 2:>rinci})al named varieties. 

 Some of the intermediate forms and deviations from the type are 

 mentioned in the lists of specimens which J have examined. The 

 measurements of the corolla given in the keys of the Parvijlorce and 

 Grandijlorcp are those of Wettstein, but, as already stated, these 

 measurements do not always aj^ply to British plants. 



Kkv to the Bkitish Species. 



Up])er cauline leaves, excluding the few patent 



teetli at least twice as long as broad ; 



capsule glabrous or with a few decumbent 



hairs on the margin 19. E. sali><hiir(j- 



L'pper cauline leaves at most twice as long as [cni;is. 



broad ; capsule with ei'ect hairs on the 



margin I, 



