MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 427 
it appears (cf. p. 337) that the ternary system Diops:Ab:An 
must form two melting-planes, one for Diops and the other for 
Ab+An. This is verified by Bowen’s experimental investiga- 
tions, whereby further is proved that the eutectic line between 
Diops, 1394 
Ab, ca 1100" ~— An,1550° 
Fic. 6.—The ternary system An: Ab: Diops (horizontal projection), after Bowen 
Epiops-An. 2Nd_ Epjops—ap. has a continuous decline, without a maxi- 
mum ora minimum. From Bowen’s diagram (Fig. 6) I calculated 
some points on the eutectic line. 
Melting-Points 
58 Diops:42 An: o Ab or 58 Diops: 42 AnyooAbo 1270° 
51 Diops:39 An:10 Ab or 51 Diops:49 An gAbz ca. 1260° 
44 Diops:36 An:20 Ab or 44 Diops:56 An ¢;Ab,;; ca. 1238° 
38 Diops: 32 An:30 Ab or 38 Diops:62 An »Abs ca. 1235° 
33 Diops:27 An:40 Ab or 33 Diops:67:An pAb ca. 1225° 
28 Diops: 22 An:50 Ab or 28 Diops:72 An jAby ca. 1215° 
23 Diops:17 An:60 Ab or 23 Diops:77 An2Ab;3 ca. 1205 
18 Diops:12 An:70 Ab or 18 Diops:82 An;;Abs; ca. 1185° 
The sequence of crystallization is illustrated by some examples 
taken from Bowen’s treatise, to which, however, I add a few 
comments. In a melt (m) of 50 per cent CaMgSi.O, and 50 per 
cent Ab,;.An;,—accordingly with a surplus of CaMgSi,O.—diopside 
