70 THE JOrRNAL OF EOTAT^T 



Tlmhellatce, As^perifolicp, Siliqvosce and the like. A. L. de Jussieu 

 (Gen. PI. p. Ixxii ; 1789), on the other hand, employed many plurals of 

 typical genera (or old generic names) — e. g., Junci, Onagrce, Nycta- 

 qines, Polemonia. Such names were afterwards felt to be inappro- 

 priate, as they should denote, strictly speaking, only the species of 

 Juncns, Onagra, Nyctago, and Folemonium respectively. A. P. 

 De Candolle (Theone Elem. 213 ; 1813) accordingly adopted the 

 following sutKxes which had been used by various authors to indicate 

 that the families were composed of plants related to their type 

 genera : -acece, -icece, -ecB, -inece^ -anece, -arieoe, -idece^ e. g., Tiliacecd, 

 Hippocraiicece^ MenispermecB, Cistinecs, Flacoiirtianece (ed. 2 ; 

 1819), Onafjrariace, Folemonidece. The actual form given in tlie 

 Theorie Elementaire was gallicized in accordance with prevailing- 

 custom in books of a semi-popular nature. The corresponding Latin 

 forms were mostly given by the authorities cited b}^ De Candolle. 

 The choice of the particular suffix was largely a matter of euphony. 

 Such diversity of termination for groups of the same rank was found 

 to be confusing, especially for teacJiing purposes, and Lindley (/. c.) 

 accordingly advocated the uniform adoption of the suffix -acece for 

 family names. 



It is, I think, clear from this brief historical sketch that the 

 priority of a famil}^ name formed from that of a genus should date 

 from its first publication with a diagnosis, notwithstanding that it 

 may have appeared in the form of the plural of the genus or with 

 some suffix other than -acecB. In such cases the name of the original 

 author should be cited in parentheses. Whatever rule is adopted, 

 however, it will be necessar}^ to have a list of nomina conservanda 

 for families, in order that well-known names ma}^ rfot be superseded 

 on purely technical grounds. 



A few examples may now be considered of family names which 

 should be accepted under the International Kules. 



The name Scheuclizeriacece- has been adopted in recent years by 

 some authors in place of Juncaginaceco (Buchenau in Engl. Pflanzen- 

 reich, iv. 14 ; Engl. Syll. ed. 7, 120). But Jiincaginacece is valid 

 under the Kules. It was formed from the old generic name Juncago 

 Tourn. (adopted in Moench, Meth. 644 ; 1794), a synonym of Tri- 

 glochin, and was published in 183(3, whereas ScheiichzeriacecB dates 

 onlv from 1858. The forms Juncagines and Juncaginece are still 

 earlier. Incidental h^ it may be mentioned that Scheuchzeriacece 

 Agardli was a segregate from TriglocliinecB and included only 

 Scheuclizej'ia. 



JuxcAGiXACE.T. (L. C. Rich.) Liudl. Nat. Syst. ed. 2, 367 (1836). 



Juncagines L. C. Rich. Anal. Fruit, p. ix (1808). 



Juncaginece L. C. Rich, in Mem. Mus. Par. i. 365 (1815) ; 

 M. Micheii in DC. Monogr. Phan. iii. 94. 



TriglocJiinece Dumort. Anal. Earn. 59, 61 (1829) ; Agardh 

 Theor. *42. 



Scheuchzeriacece Agardh Theor. 44 (1858) ; Buchenau in Engl. 

 Pflanzenr. iv. 14, 1. 



LilceacecB Hieron. in Ber. Ges. Naturf. Ereunde Berlin, 1878, 

 116. 



