162 T. Crook—Titaniferous Volcanic Rocks. 
in the rocks of any region should be made a fundamental feature 
in characterising a petrographic province. At any rate, his statement 
that ‘‘the higher figures found in the table are unparalleled else- 
where, except in titaniferous ores,’’ seems not to be in accordance 
with facts. 
Another feature of Dr. Washington’s preliminary notice appears to 
make a somewhat heavy demand upon our credence. The analyses 
recorded, as already mentioned, are of specimens from only four 
localities, two of which are small adjacent islands, but the other 
localities are widely separated; and we are asked to regard these few 
analyses of certain widely separated rocks as indicative of a condition 
prevailing throughout that portion of the Western Mediterranean 
region which he has partially defined. 
In this connection, one important fact already mentioned should be 
specially emphasised, and that is, that no mineral exhibits a more 
pronounced tendency to segregation than does titanic iron-ore. That 
such is the case has been abundantly proved. In some cases, whole 
mountains are practically built up of material which has been segregated 
in this way. Such segregated titanic ores are usually local, and their 
occurrence is significant as indicating that titanium dioxide has an 
irregular distribution in rock magmas, using the term magma in 
a broad regional sense. Even the method of sampling would appear 
to be an insufficient correction for such an irregular distribution, as 
the segregations are likely to be either too prominent or altogether 
beyond observation. 
This brings us to the important question: What are the criteria of 
a comagmatic region? If such criteria are to be scientific rather than 
merely empirical, it would appear important that they should be 
based essentially on features of somewhat persistent and uniform 
developmerit. In this respect, varying amounts of constituents such 
as alkalies may be admitted to a high place as criteria, for not only do 
they seriously affect the constitution and physical properties of 
essential minerals; they are also partly responsible for the production 
of a characteristic group of minerals of great importance, and they 
have a somewhat uniform distribution. 
But the same claim can scarcely be urged for titanium dioxide, since 
it separates early, is Hable to segregation, and does not vitally affect 
the chemical constitution and physical properties of any important 
group of the minerals according to which rocks are classified. It 
remains to be seen whether further work in the Western Mediterranean 
area will justify the use of titanium dioxide as the principal index of 
a special province. The evidence so far presented is far from being 
conclusive. It would appear that Dr. Washington has laid too much 
stress on the significance of the titanium content of the few rocks he 
has so far analysed, and it remains to be demonstrated whether even 
these are typical, for the area is a large one. Further, it is a well- 
known fact that the estimation of titanium 1s a matter of considerable 
difficulty, and even where it has not been altogether neglected the 
stated percentage is often untrustworthy, being in many cases probably 
much too low. ‘This fact adds considerably to the necessity of 
exercising caution in arguing from a comparison with hitherto 
published analyses. 
