Jan., I92i] CHASE — THE LINNAEAN CONCEPT OF PEARL MILLET 
45 
Panicum glaucum is here composed of: (i) The diagnosis given in the 
Systema (part of it applying to pearl millet and part to yellow foxtail), 
followed by Fl. zey. 44" (pearl millet only). (2) The Gronovian diagnosis 
and reference, without "7" (applying to yellow foxtail). (3) ''Habitat in 
Indiis & Italia" (notwithstanding the reference to Gronovius, Flora vir- 
ginica). (4) The description verbatim from P. glaucum in the first edition 
(the statement that the bristles are the length of the flowers, and the observa- 
tion of the flowers disposed in series, applying conclusively to pearl millet), 
with the addition ''Semina striis undulatis notata" (applying to the fruit 
of yellow foxtail and not to that of pearl millet) as in the Systema. 
Holcus spicatus (page 1483) is composed of: 
Holcus glumis bifloris muticis, floribus geminis penicillo involucratis, spica 
ovato-oblonga [the diagnosis from the Systema, 1759]. 
Panicum spica simplici aequali, pedunculis bifloris. Roy. liigdh. 54 [Un- 
identifiable by the writer, may well be pearl millet]. 
Panicum indicum, spica obtUvSa caerulea. Bauh. pin. 7. tJieatr. 522. [The 
second citation under P. americanum in the first edition of Species 
Plantarum. The *'theatr." referred to is the illustrated Theatri 
botanici, 1658. The figure is copied by tracing from that in Clusius 
for P. americanum, and does not represent pearl millet. The long 
description says among other things that the culm near the base is of 
an elegant blue and shining purple and the pith spongy, characters that 
suggest sorghum. But the further description of the culm as dividing 
into branches does not apply to sorghum. (Was it perhaps the 
illustration that was described?) The spike is said to be sometimes a 
finger long and sometimes only an inch, and to resemble "Frumenti 
Turcidi" (maize). It is said to be from the Indies and also from 
Peru, to be grow^n in gardens in Belgium, rarely in Germany, from seed 
sent from Spain. Altogether it reads like a compound of half-re- 
membered plants of sorghum, common millet, and maize. Such a 
figure and description at any rate can not reasonably be taken as a 
basis for a name.] 
Panicum americanum. Clus. hist. 2. p. 215. [See note on P. americanum 
above.] 
Gramen alopecuroides indicum maximum. Raj. hist. 1908. [Ray's de- 
scription applies very well to pearl millet.] 
Gramen paniceum sylvestre maximum indiae orient. Pluk. aim. 164 
[error for 174] /. 32. /. 4. [The figure, which was referred to under H. 
spicatus in the Systema, may well be pearl millet.] 
Habitat in India. 
Culmus bipedalis, crassitie pennae cygneae, tectus vaginis foliorum hispidis 
ut ipse culmus. Folia saepius 10, latitudine digiti hispida. Spica 
crassissima pedicellis brevissimis apice fasciculo setarum, intra quem 
Flores 2, sessiles. Calyx bivalvis, membranaceus, biflorus. Petalo 
exteriore hermaphroditi mucronato; masculi obtuso. Stylus floribus 
longior, lanatus, laeviter apicae bifidus. Antherae oblongae. [A 
hispid culm two feet tall, and as thick as a swan's quill, covered with 
hispid sheaths and with hispid blades, is certainly not pearl millet, 
but the rest of the description might apply to it, or, somewhat better, 
to common millet. Did Linnaeus possibly have a sterile plant of 
