62 PROCEEDINGS OP THE AMERICAN ACADEMY 



almost exclude from Ericacece. Much preferring to retain the order 

 as a jireat whole, I would combine Clethra and the tribe Fyrulece into 

 the third suborder, Pyrolinece. 



Galax aphylla, Linn. The name first appears in the first edition 

 of the Species Flantarum, p. 200, in 1753. Neither generic nor spe- 

 cific name has any fitness ; the herb is not milky, nor is it leafiess, 

 except as to tiie scape. The plant in view, the Anonymos s. Belvedere, 

 of Clayton, is recognizable by the good generic character in the first 

 edition of the Flora Virginica, and from Clayton's herbarium. The 

 generic character of Galax first appears a year later, in the 5th 

 edition of the Genera Plantarum, and it is, as has been noted, utterly 

 incongruous with Clayton's plant, to whicli Linnasus meant to apply it. 

 This generic character Linnasus copied from Mitchell's Nova Planta- 

 rum Genera, viz., from his VUicella, merely substituting the name of 

 Galax. Consequently not a word of the Linn;«an generic character 

 is applicable to Galax aphylla^ Linnanis's only species ; wherefore it is 

 not surprising that Andrews, Richaid, and Ventenat should have re- 

 spectively described that plant under other generic names. Although 

 the contradiction was long ago pointed out, still most authors, down 

 to Endlicher, De Candolle, and later, have followed LinuiBus in citing 

 Viticella as a synonym of Galax. Gronovius, in the second edition 

 of the Flora Virginica, was evidently struck by this total discrepancy ; 

 and he covered it in a curious way, by omitting altogether the correct 

 character of Clayton's plant, as printed in the original edition. It was 

 reserved for INIr. Bentham to divine what Mitchell's Viticella really is, 

 viz., HydropJtyllum appe^vliculatum, to which the name of Galax etymo- 

 logically is equally inap|)lical)le. See lk'ntli.& Ilook. Gen. ii. 827. 



STEIRONEMA, Raf. in Ann. Gen Phys. Brux. vii. 192 (1820). 

 Genus between Trientalis and Lysiinacliia, distinguished from both by 

 the presence of staminodia (the rudiments of the other series of 

 stamens) between the fertile filaments, and by the {estivation of the 

 corolla, in whicli each division is separately involute around, or even 

 convolutely enwraps the stamen before it. The latter character, which 

 I have I'ecently ascert dned, is not alluded to by Baudo in his index of 

 the caulescent Anagallidece (Ann. Sci. Nat. ser. 2, xx.), nor by Bentham 

 and Hooker in the second volume of the Genera Plantarum, in which 

 tlie aestivation of the corolla is first systematically einployed in the 

 arrangement of this order.* Following Bigelow (§ Seleucia) I had 



* The tribe Li/simackiem is characterized as having convolute (or " contorted ") 

 aestivation of the corolla; the Primulece, by quincuncially imbricated. This 



