4 University of California Publications in Agricultural Sciences [Vol.3 



ciliato constans. Gluma V (fertilis) 3.5 mm. longa ovato-lanceolata, acuta, cliar- 

 tacea, appresse pubescens, tenuissime 5 nervis. Palae glumam aequans, angustior, 

 carina ciliolata. Antherae 3.5 mm. longae. Ovarium glabrum. Caryopseos macula 

 hilaris fere dimidiam caryopsin aequans. 



Patria ignota, culta in Australia sub nomine Phalaridis commutatae. Plantam 

 et semina misit A. J. Ewart, Melbourne. 



Es ist auffallend, dass diese gut unterscliiedene Art, welehe in Australien als 

 Futtergrass gebaut und sehr geriihmt wird, bisher meiiies Wissens nirgends 

 beschrieben wurde. In Australien wurde sie durch Mr. Harding, Kurator des 

 Botanischen Gartens in Tuwumba, Queensland (unbekannt woher) under dem 

 Namen Ph. coinmutata eingefulirt und unter diesem Namen von Samenhandlern 

 in Melbourne verbreitet. Ein mir vorliegendes Eeklameblatt zeigt die Darstellung 

 eines dichten Kasens van angeblich 7 Fuss (2.2 m.) Hohe, der nach dem Schnitt 

 in 46 Tagen wieder eiuen 41 Zoll (106 cm.) holien Easen hervorgetrieben hatte. 

 Besonders wird sein wert als Wintergras hervorgehoben. 



This excellent detailed description agrees with our grass from 

 South Africa in everything but the sterile florets. As these are used 

 as the chief distinguishing characters in the genus to separate one 

 species from another, a disagreement in regard to these particular 

 structures makes a positive identification difficult. Our specimens 

 show a variation in the sterile florets, one 1.5 mm. in length and 

 the other much smaller, .7 mm. The latter may be reduced to a mere 

 point protruding from the ovate scale (pi. 1, fig. 4). 



Hackel's description is as follows: 



Gluma III nulla, IV vacua 1 mm. longa . . . infra apicem squamulae inserto 

 apice penicillato-ciliato constans. 



That Hackel seemed convinced that there was constantly only one 

 sterile floret is emphasized by the fact that in a discussion of the 

 relationship of the new species, he writes "doch ist der Hiillspelzen- 

 fliigel bei Ph. stenoptera noch schmaler als bei Ph. nodosa und es ist 

 stets nur eine kleiue Leerspelze (die glume IV) am Grunde der Vor- 

 spelze"; also "Ein merkmal aber, das sie von beiden genannten Arten 

 (Ph. arundinacea and Ph. bulbosa) scharf trennt, ist das Fehlen der 

 gluma III, das mir ganz Konstant zu sein scheint. " 



Through the kindness of Mr. Harding, who forwarded us some old 

 seed from the Botanic Gardens at Toowoomba, Queensland, we were 

 able to examine original material. There were present in the packet 

 some spikelets with two sterile florets, others with one sterile floret, 

 ;iikI another very minute one and still others with only one present. 

 r l'h is same condition was found in our specimens grown at Davis and 

 which may be seen under sheets nos. 5000, 5001, 5002, and 5003 of 



