1 248 PANJAB CASTES.
2 " Asia. In the Paniab they arc so nnmeroua that thoy cannot all be rich and mercantile ; and
3 '■ many of them hold land, cultivate, take service, and follow various avocations, i
4 " The Khatris are altogether excluded from Brahmin Kashmir. In the hills however the
5 " 'Kakkas,' on the cast bank of the Jahlam, arc said to have been originally Khatris (they are
6 " a curiously handsome race), and in the interior of the Kangra hills there is an interesting
7 " race of fine partiarchal-lookiug shepherds called Gaddis, most of whom are Khatris. Khatri
8 "traders arc numerous in Dclili; are found iu Agra, Lucknow, and Patna ; aud are well known
9 " in the Bara Bazaar of Calcutta, though there they are principally connected with Tanjab
10 " firms.
11 " The Khatris do not seem, as a rule, to reach the western coast : in the Bombay market ^ I
12 " cannot find that they have any considerable place. In Sindh, however, I find in Captaiu
13 " Burton's book au account of a race of pretended Kshatriyas who are really Bauias of the
14 " Nanak-Shahi (Sikh) faith and who trade, and have a large sliare of puljlic oftices. These are
15 " evidently Khatris. Ludhiana is a large and thriving town of mercantile Khatris, with a
16 " numerous colony of Kashmiri ahawl-weavcrs."
17 Within the Panjab the distribution of the Khatri element is very well
18 marked. It hardly appears east of Ludhiana, the eastern boundary of the
19 Sikh religion, nor does it penetrate into the eastern hills. It is strongest in
20 the central districts where Sikhism is most prevalent, and in the Rawalpindi
21 division and Hazara, and occupies an important position in th(! western Hill
22 States. Although the Khatris are said to trace their origin to Multan, they
23 are far less prominent in the southern districts of the Western Plains, and
24 least of all on the actual frontier j but this would be explained if the Aroras be
25 considered a branch of the Khatris.
26 As Sir George Campbell remarked, it is curious that, intimately connected
27 as the Khatris always have been and still are with the Sikh religion, only 9
28 per cent, of them should belong to it. Nor do I understand why the pro-
29 portion of Sikhs should double and treble in the Jahlam and Rawalpindi
30 districts. Some i,M)0 are Musalmiln, chiefly in Multan and Jhang where
31 they are commonly known as Khojahs ; and these men are said to belong
32 chiefly to the Kapur section. The rest are Hindus.
33 540. The divisions of the Khatri Caste. The question of the sub-divisions of the Khatris is
34 exceedingly complicated. Within recent times tlierc has sprang up a system of social gradua- [p. 296]
35 tiou in accordance with which certain Khatri tribes refuse to intermarry with any save a certain
36 specified numlicr of their fellow tribes, and the distinctions thus created have been formulated in a set
37 of names such as Dhaighar, " he who only marries into two and a half houses ; " Chdrsati, " he who
38 marries into four tril)es ;" Chhezdti, " he who marries into sL\ tribes ;" and so on. This purely arti-
39 ficial aud social classification has obscured the original trilial divisions of the caste ;^ for Khatris of
40 the same tribe may be in one part of the Province Charziltis, and in another Barazatis and so forth .
41 It has also terribly confused tlic entries in the schedules, assi-ted by an unfortunate mistake in
42 the sample schedules issued with the instructions to enumerators, in which, owing to my own
43 ignorance of the matter, one of the pancliai/ati or artificial divisions was shown as a tribe. The
44 distribution of the main sections is shown in Abstract No. 92 below*. It will be noticed that *p_ 249
45 they include more than three-quarters of the total Khatris of the Province, hut that the percent-
46 ao-e unclassified is very large in some districts. In others again the number classified is larger
47 than the total Khatri population. This is due to the same figures being in some_ cases repeated
48 twice over. Thus in Giljranwala 963 Khatris have returned themselves as Kapiir Charzati, and
49 80 appear under both heads ; and so in other cases also.
50 The headings of the .\bstract include three different kinds of divisions, first the four real
51 tribal sections, tlien the four most important of the artificial divisions alluded to above, and fina.ny
52 six of the most important clans. The origin of the division into the four sections called Bun;iahi,
53 Sarin, Bahri, and Khokhran, is said to be that Ala-ul-diu Khllji attempted to impose widow-mar-
54 riage upon the Khatris. The Western Khatris resolved to resist the innovation, and sent a
55 deputation of 52 (hdwan) of their members to represent their case at court ; but the Eastern
56 Khatris were afraid to sigu the memorial. They were therefore called followers of Shara A'j'.n.
57 or the llahomedan customs — hence Sarin — wliile memorialists were called .Bawan^ai from the
58 number of the deputation or of the clans respectively represented by the members of the deputa-
59 tion ; hence Bunjahi. The Khokhran section is said to C(jnsist of the descendants of certain
60 Khatris who joined the Khokhars in rebellion, and with whom the other Khatri families were
61 afraid to intermarry ; and the Bahri section, of the lineage of Mahr Chand, Khan Chaud, and v