222 THE JOURNAL OF BOTANY 



" Caules plures undique procumbentes depress!, digit! longitudine. 

 Foliola pinnata, foliolis 11 s. 13 lanceolatis, vix manifeste pubes- 

 centibus, acutis. Capitula florum axillaria, subsessilia. Legumina, 

 saepius 8, ovata s. cordata, depressa, deorsuin parum complicata, 

 acuta, incana, apice adscendentia, ceraua, rigido pedicello." 



At last we reach the Mantissa altera, where, on p. 274, 

 A. pentaglottis is defined as "A. caulescens procumbens, legu- 

 minibus capitatis replicatis compressis conniventibus cristatis 

 acumine reflexo," with a reference to Herm. tab. 75, and to the 

 " Glaux hispanica " of Eivinus. " Habitat in Hispania." 



A. epiglottis is defined as " A. caulescens procumbens, legu- 

 minibus capitatis sessihbus cernuis cordatis mucronatis repHcatis 

 nudis," with the synon^^ms " Astragalus siliquis cordatis acutis 

 pendulis, lateribus conniventibus. Hort. Cliff. 362." " A. his- 

 panicus, siliqua epiglottidi simili, flore albo minor. Herm. Lugdb. 

 76, t. 77," and " Glaux ininima Eiv." " Habitat in Orientis, 

 Hispaniae, Galloprovincise sylvis montosis. Exclusus in Sp. PL 

 1069 ita restituatur. Corollse albse." 



Then A. hypoglottis. "Astragalus caulescens prostratus, 

 leguminibus capitatis ovatis replicatis compressis pilosis acumine 

 reflexo. — Astragalus epiglottis Syst. Nat. 12, p. 199 [499] exchisis 

 synonymis. — Astragalus villosus procumbens, floribus pallide 

 purpureis oblongis. Eaj. App. 454 [455] . Habitat in Hispania. 

 Similitudinem gerit cum A. pentaglotto, sed magis caulescens. 

 Capitula pedunculata pedunculo similiter elongato ; Flores plures 

 ad 8 s. 10, in capitulo rotundato more trifolii, c^rulei aut pur- 

 purascentes. Legumina replicata, ovata (non subulata), acumine 

 duplici (in maturis), subulato, recurvo ; tecta pilis (non squamis 

 furfuraceis) longis, albis, mollibus, minime autem in orbem expansis 

 quemadmodum in A. stella, neque subulata ut in A. sesameo." 



The new synonym here introduced from Bay's appendix is too 

 uncertain to give any help. Bay, after the diagnosis, " says, 

 " Siliqua uncialis est, et longior, bicapsularis, rotunda, magnitudine 

 Loti hasmorrhoidalis [i. e. Dorycnii hirsuti) in acumen desinens. 

 Ex horto Amsterod. a Breynio missus." I have no notion 

 what Bay's plant can be. It does not fit any of the species 

 which have been suggested for A. hypoglottis. The legumes of 

 Dorycnium liirsutum are far from being " iinciales," but they do 

 not differ much in size from those of Astragalus purpureus, which 

 has distinctly larger legumes than .4. danicus, and much larger 

 ones than A. asperulus Duf. or A. epiglottis. 



We must now consider the herbarium specimens in the light 

 of what has been said. These are seven in number ; five in 

 Linnaeus's own herbarium, and two in that of the Hortus 

 Cliffortianus. 



(1) A specimen from the Hortus Clijfortianus of true .4. epi- 

 glottis, labelled thus : — 



* Ray attributes this diagnosis to Sherard, but it is not to be found among 

 the Astragali in Sherard's Schola Botanica, 



