S.Uj I UK .luLUNAL 01- HOIANV 



tJir(?i broad; upper part of stem aiul branches o£ inflorescence with 

 numerous strong patulous setae and frecjuent glands shorter than the 

 seta?. InHorescence congested even at the end of flowering. Calyx seg- 

 ments subpatent at tbe apex, scarcely exceeding the throat of the corolla. 

 Fruiting calyx ampliate at the base, campanulate. Limb o/* unfolded 

 corolla jieir if or m, dull violet. Tabe quite ///«6r6>?^s below the ring of 

 hairs in the throat. 



The British plant has the leaves normally spotted, the radical ones 

 exactly as in Kerner's Kgure of P. longifolia, the caulineones normalh^ 

 patulous, 4-7 times longer than broad ; the flower certainly not azure, 

 but purplish blue, pelviform and not campanulate (in Kerner's usage of 

 the terms) with a glabrous tube below the ring of hairs in the throat. 

 It is therefore diflicult to see how Kerner could have avoided identi- 

 fving our plant with his P. longifolia had he known it. 



With regard to the presence or absence of spots he is very decided. 

 He says (p. 11) '' the foliage leaves in the true P. tuberosa of 

 Schrank are always unspotted. Those authors who state that 

 P. tuberosa also has spotted leaves either do not know the true plant 

 of Schrank, or have mixed several species under this name." Again 

 (p. 16) : — ''They [Grenier and Godron] therefore believe that both 

 of these species [P. aziirea and P. tuberosa^ may have spotted leaves, 

 which I deny ; for neither the true P. tuberosa of Schrank, nor 

 P. anc/usti folia L. Fl. suec. have spotted leaves, and spotted specimens 

 of ' P. coigustifolia ' and of ' P. tuberosa,'' of which Gren. and Godr. 

 implicitly speak, are without doubt the narrow and broad leaved 

 forms of P. longifolia Bast." 



It seems evident that Kerner had not seen British specimens. 

 His only mention of the British plant is in the account of the 

 distribution :— " Amjlia. Ins. AVight, Hants and Dorset, Hampshire 

 [s/c]." This suggests mere book knowledge. He may have relied 

 on the figure in English Botany, where the leaves are figured without 

 spots: the text, however, says "much less spotted" than those of 

 P. officinalis. The specimens in Herb. Sowerby have the radical 

 leaves conspicuously spotted, the stem leaves evidently, though 

 less, so. Bromfield, Syme, and Babington all refer to the spots ; only 

 F. X. Williams, who names it P. azurea, omits mention of them. 



By the kindness of Dr. Sharp and Miss Piftord I have received 

 large fresh series of the plant from Brockenhurst. The spotting is 

 very variable. Most commonly three or four rather large spots on 

 each side of the midril) ; sometimes very ninnerous to few small ones are 

 also scattered over the leaf. The spots are quite connnonly absent, 

 but may vary in number on the same plant. They ai'e usually jjale 

 green, but are sometimes, as in the Sowerby s])ecimens referred to, 

 very consi/icuous and white. Bromfield says " the leaves are usually 

 nebulously spotted with greenish white, more rarely quite plain ; 

 sometimes these spots are very large, and confluent, occupying the 

 greater portion of the leaf." This last })hrase would fit P. saccharata 

 exactly, and such confluence has so far not been noted except for this 

 and (rarely) the closely allied P. affinis. But some of a fine series of 

 plants sent from near Wimborne (Dorset) by Miss V. M. Dale 

 agree exactly with this description. The leaves of one plant were 

 unspotted ; all the rest had large white spots ; two were white over 



