364 
The N.Z. Journal of Science and Technology. [Dec. 
second 9 because it is Zoology, the third 4 because it is Mollusca, the fourth 
3 because it is Gastropoda. This is as far in zoological classification as 
the system has been elaborated. The added numbers 931 indicate the 
geographical distribution, New Zealand. 
In the International Catalogue the paper would be included in a volume 
of the series N, Zoology, and in section VIII, Mollusca. It would appear 
in the subject index first under 2227 ik , Distribution, (a) Recent, (f3 ) Terres¬ 
trial, New Zealand, and again under 2231, Taxonomy, under the further 
printed headings Gastropoda, Taenioglossa, Hydrobiidae, Potamopyrgus. 
In the system of the Concilium Bibliographicum there would be cards 
each of which would bear in the bottom left-hand corner the decimal number 
59, indicating Zoology. On the top right-hand corner one card would bear 
the notation 19 (931) 4*32 Potamopyrgus. 19 signifies geographical 
distribution, (9) Oceania, (93) Australasia, (931) New Zealand, 4 Mol¬ 
lusca, 4”3 Gastropoda, and 4*32 Prosobranchia. Another card would 
bear the same notation in a different order, but omitting the 19—viz., 
4 Potamopyrgus (931). The cards would be indexed strictly according to 
numbers, commencing with the left-hand digits, so that the first card would 
fall into a group forming a complete bibliography of the fauna of New Zea¬ 
land, and the second into a taxonomic index of papers on all species of the 
genus. There are further numbers which may be used to connote other 
zoological viewpoints than distribution and taxonomy, and they may be 
combined in various ways. The system is complicated, but exceedingly 
comprehensive, and is perhaps the best devised for the sciences which it 
treats. 
It is obvious that the last two bibliographic systems are unsuitable for 
purposes of library classification, but that the systems of library classifica¬ 
tion are unlikely to be sufficiently detailed for the purposes of a subject 
index such as required for the scheme of co-operation which I have suggested. 
Since the Dewey system is already in use in most of the larger libraries in 
New Zealand, it will probably be best to adopt this system for library 
classification in the central library, unless the system of the Institut Inter¬ 
national proves to be much superior. On the other hand, it may be desirable 
to adopt for the subject index an extension of the Dewey system by a 
series of arbitrary numbers .ordetters, as is done in the Concilium Biblio¬ 
graphicum. I prefer to keep an open mind on this subject until I have 
seen the manual of the Institut International, and in the meantime in the 
suggested card have made provision for both notations. 
It is obvious, however, that for serials containing a number of papers— 
and the library must consist mainly of such books—no subject index entry 
for the whole volume is practicable. The inquirer—-say, in Dunedin—finds 
out by some means or other that a paper he wishes to consult is in the 
Ann. Chim. Analyt., Paris, vol. 19, 1914. All he wishes to know is where 
he may find that volume, and that the author’s index at once informs him, 
or lets him know that there is no copy in any New Zealand library. 
The Library Building. 
The whole of the above scheme is dependent in the first place on the i 
provision of a suitable fireproof building. This should contain a well-lit 
public reading-room, a large stack-room also open to readers, small private 
reading and working rooms for Government officers and approved research 
workers who have to do much reading, working-rooms for the staff (in¬ 
cluding printing and binding rooms), and storerooms to hold duplicate 
books and the stocks of Board of Science and Art and other publications 
