THE GENUS PUMARIA IN BRITAIN 23 



what dorsally compressed hut not broad, apictUate, with a spathu- 

 late dilation of the wings (exceeding the keel and nearly reaching 

 the apex), w^iich is apparent in the hud stage, but is obscured 

 when the wings reHex upwards in flower and cover the keel ; 

 spur about as long as the sepals. Loioer petal luith erect and 

 narrow margins. Fruits very small, not exceeding 2 mm. long 

 (including the apiculus), and somewhat less in breadth, subrotund- 

 ovate in profile, and subacute or apiculate, even when mature and 

 when dried ; little compressed laterally, and obscurely keeled, with 

 the fleshy neck almost obsolete ; smooth when dry, with small 

 and faint apical pits. 



p. LOWEi, nov. var. (not known as British). 



Exsiccata. — Lowe, Port Orotava, Nov. 16, 1857, Hb. Mus. Brit. ! 



Differt a typo racemis saepe multifloris (sub-20 fl.), petalo 

 superiore alls apicem attingentibus obtuso, fructibus sine apiculo 

 obtusis subglobosis. 



Differs from the type in its longer and more floriferous 

 racemes, its blunt upper petal, toith wings reaching the apex, and 

 its obtuse and subglobose fruits. 



y. DECiPiENS, nov. var. 



Habitu foliisque typo similUma sed pedicelHs recurvatis vel 

 flexuosis, sepalis 4 mm. longis 2^ mm. latis breviter acuminatis, 

 corolla 11 mm. longa, fructibus typi forma sed certe majoribus 

 cum apiculo 2| mm. longis. 



In habit and foliage resembling the type, but with flexuous, 

 and sometimes recurved pedicels. Sepals 4 mm. long and 2| mm. 

 broad, shortly acuminate ; corolla 11 mm. long ; fruits in form 

 like the type, but distinctly larger, 2^ mm. long, including the 

 apiculus. 



Distribution. — F. muralis (sensu stricto) is a rather rare and 

 local species in Europe, but it appears to be common in Madeira 

 and the other Atlantic Islands. A fragmentary specimen at the 

 Herb. Mus. Brit, that I think can be referred to no other species 

 is labelled " St. Helena, Cook's First Voyage." 



It is also locally abundant in South Africa, and Harvey's 

 statement in the " Flora Capensis," where it is recorded as F. 

 officinalis var. capensis, that it was introduced from Europe seems 

 exceedingly doubtful. If not native, it is much more probable 

 that it was at some time brought from Madeira or the Canaries. 



It occurs in Norway [Hb. Keio)\ Germany! Holland [Nyman), 

 France! Spain (JJ^. ^ew) ! Portugal! Algeria (H'a?6S5 A:.). Azores! 

 Madeira! Canaries! Ascension! St. Helena ! South x\frica ! Ber- 

 muda ! Brazil [Haussh.). 



F. muralis was formerly recorded for many counties in Ireland 

 and is given in Nyman's Conspectus, " Hibernia, pr. Belfast, 

 Corry, 1882." iVllthe Irish specimens that I have seen, however, 

 belong rather to F. Borcei, with the doubtful exception of one, 

 which unfortunately shows no fruit, collected at Portadown by 

 Miss McArdle in September, 1904. 



