the same as that of the Big Valley, although the natural way of communications is through the Small Valley and a series of passes in the mountains:

Besides there is all the series of the duals, trials, etc.:

(incl.) or (excl.)

"we two" iropanu "we three" igyonanu "we four" ida panu

(incl.) or (excl.) iropadu "we five" idigayanu idigayadu igyonadu "we six" idicopanu idico padu ida padu "we seven" idigajenu idigajedu "we all" iranu iradu

Note in this respect that the first four show a certain amount of contraction; after that they can hardly be called "formal" plurals. Here are the Tehuantepec numbers: 1 tuvi 2 tcupa 3 tcona 4 tapa 5 grayo 6 copa 7 gaje.

Dialect of the Northern Mountains

Enough has been said to allow us to treat now the dialects of the Northern Mountains in a very rapid manner and almost without comments. It will be seen how different they are, and yet built on the same basis:

1) Sub-dialect of the Basin of Ixtlan

The second plural needs an explanation. The Macuiltianguis form is clearly a pluralized polite address. "Your honour," "Usted" is found almost everywhere in the Zapotecan region. It appears as yubyu in the Valley, and kwina or kwinalu in the Sierra. The other two forms levi and livii have all the earmarks of being influenced by the Spanish use of the third for the second.

Xaltianguis: "to eat" (Rad. -o-; Incip. pfx. gh-)

All the above for "object mentioned"—if object is not actually mentioned then it would be: ghoyavi, ghoyaye, ghoyava, ghoyana respectively for any one, an older man, an animal, a thing. The -na form is of course the most frequent; it corresponds exactly to the -ing of the Little Valley; it appears as -ni in some localities, and gets generally mixed up with the element ni "to be essentially."

"The furthermost Zap. pueblo.

"This apparent anomaly is found everywhere in the Sierra, and is very useful as an index of dialectical variations, but we cannot go into it here.

It is well to observe that in Laxopa and Cajonos where the two keystone-words "person" and "animal" have been leveled by a certain. trend of Zapotecan phonetics to an identical sound be differentiated only by tone (high for person, low for animal), the more usual way of referring to an animal is to say be yijr "animal-brush" (i.e. wild animal). Another cause of complication is the fact that the demonstrative of the third singular "this" is (as in several other dialects) le with a very open and very short e, almost an a.

After these previous remarks, and remembering the strongly rooted u ending for the second singular in all dialects, the following conjugation may perhaps be understood, especially if one keeps in mind the Teotitlan forms for "object not mentioned, but it is a thing."

There are in this region (Basin of Yalalag) some localities where there has been an extremely interesting interlocking of distinct but superficially similar forms, with the result of confusion, and the passing of one form bodily over to the next. However, space forbids going into this.

(To be concluded)

NOTES AND PERSONALIA

Members of the Society will be glad to learn that the Indonesian collection of the late C. E. CONANT is not to be dispersed but will remain, as its self-denying founder would have wished, available as a unit for the use of scholars. Its home is to be the Library of Williams College to which it has been presented by an alumnus of that institution, MR. HERBERT H. LEHMAN of New York.

The collection, which includes some sixty of the Indonesian languages and dialects, contains an immense amount of material on the Philippine languages, much of which cannot be duplicated. Of its formation MR. JOEL HATHEWAY writes: 'Most of the books were purchased by him at the cost of very real sacrifice. A few were gifts from friends, Spaniards and Filipinos, fellow-enthusiasts in linguistic studies. I well remember the days, twenty-five years ago, when his collection was started; when together we visited the monasteries and bookstores of Manila and Cebu, and Conant would spend his last peso for grammars and dictionaries of the native languages. Within even a week of our arrival in the Islands he had embarked upon the comparative study of the native languages which was to be his life work.'

The first number of LANGUAGE MONOGRAPHS entitled A Technique for the Experimental Investigation of Associative Interference in Artificial Linguistic Material by ERWIN A. ESPER has now been distributed. The publication of another number Post-consonantal w in Indo-European by FRANCIS A. Wood is beginning. The Society is able to assist such publications to a limited extent, and can moreover, secure from its printer terms which seem very advantageous. The Committee on Publications will be glad to hear from any member who has material suitable for inclusion in this series.

In the next number of LANGUAGE it is planned to begin the publication of book reviews. The editor will be pleased to hear from any member who wishes to share in this undertaking. Copies of the works listed under 'Books Received' can be furnished for the purpose.

The following changes in the activities of our members may be recorded: