(lower, fupported on a fcape. It was in this form only that 

 the plant had occurred to HalLer, at the time he wrote his 

 Hijioria Stirpium Helvetia ; probably alfo Linn^us had not 

 feen it in any other, when he described the leaves as radical, 

 and called the peduncle a fcape : and even in cultivation the 

 feedling plants will fometimes flower before any running fhoot 

 appears. 



The neftaries or internal petals have been hitherto faid to be 

 numerous, but we conftantly find four that are fomewhat different 

 from the reft, placed alternately with the petals and without 

 any veitige of anthers; all the others, generally about twelve, 

 having more or lefs appearance of anthers at their tips, we 

 confider as dilated filaments. The nectaries afford an excellent 

 mark of difcrimination between Atragene aujiriaca and 

 fibirica* which are otherwife not eafily diftinguifhed by words, 

 thofe of the latter being emarginate or linear-obcordate, whereas 

 thofe of the former are quite entire at the point. 



Having had an opportunity of feeing three fpecies flower at 

 Mr. Lod dices, at Hackney, this fummer, we have not 

 omitted comparing them together, and think they may be 

 fafely concluded to be diftincl: and characterized as follows : 



1. Atragene aujiriaca; foliis binis duplicato-ternatis : 

 foliolis ovatis ferratis, ne&ariis obtufis. 



2. Atragene fibirica g foliis binis duplicato-ternatis: 

 foliolis ovatis ferratis, neclariis emarginatis. 



3. Atragene americana ; foliis quaternis ternatis : foliolis 

 cordatis integerrimis, neclariis acutis. 



Of the fourth fpecies, the ochotenfis of Pallas, we know 

 nothing but the little this author has given of it in his Flora 

 Roflica, vol. 2, p. 69. It has fix petals, but in other refpecls 

 has the habit of the reft. 



According to J uss ie u, what we have denominated corolla is 

 a calyx, and our ne&aries are petals. We prefer keeping to the 

 Linnaean terms, and have only adopted that of nectaries inftead 

 of his internal petals, in conformity to his own language in the 

 reft of the order. 



The plant now figured flowers nearly at the fame time with 

 Atragene aujiriaca^ a month later than fibirica> is hardly lefs 

 ornamental, and has the exclufive advantage of being agreeably 

 fcented. Was raifed from feeds from North-America, by 

 Mr. Loddiges ; appears to be perfectly hardy, and to produce 

 feeds freely, by which it may be propagated without difficulty, 

 and makes a very defirable addition to our climbing fhrubs. 



