Am writing this in the hope that kspv reads this since I just noticed he edited an entry in Jan 2009 (was writing this while a distraction came up as above). If kchandra comes around, hope he will read this too (please note, am not a kamma, and do not know much abt cultures followed in kamma households):
Reg K.Chandra's comments on Iyengars, please note this:
http://www.tamilbrahmins.com/showpost.php?p=17702&postcount=5 Probably those rumors were created by people who had nothing better to do than malign people based on caste just because they were successful.
Reg Kammas, so far in Andhra, population differentiation tests have showed a slight differentiation b/w these caste populations: Kapu Naidu, Kapu Reddy and Kamma. Plz refer:
http://cat.inist.fr/?aModele=afficheN&cpsidt=18906799 . Amongst the 3, only Kamma is considered as having a northindian input. Please note that a person might be of any caste, as say belonging to Kapu or Reddy, socially in the present time, but he may end up clustering more with the Kamma grouping. Similarly, a person maybe a Kamma currently socially, but may end up clustering more with Naidu or Reddy grouping (all are mixed populations). The populations of brahmin, raju, komati, kamma, kapu, reddy and lambadi in Andhra Pradesh are decidedly mixed groups.
It is understood that descendents from the previous caste system (that came about by tribal admixtures) c
ame to be scattered across various present day castes. All of these mixed populations or given individuals across these castes very likely shared common origins / a set of founding members.
After I spoke to a person, I understand this: Though perhaps the name Kamma is recent, but certain given members of this community must have existed under some other name before gene inflow happened. So, kamma must have been an old endogamous unit (otherwise population differentiation wud not show up as it has been found). No one knows what those older groups were named. And those existing groups received a northie input; and the output spread across various new-age castes (including andhra brahmin).
The person was also of opinion that though it is not known when this northindian input into native population occurred, it might have occurred a very long time ago (sufficient to allow differentiation to show up), probably during the buddhist times. And from that merger, several descendents found their way into various present day castes. For a general idea of which castes aligns more with the other, you may wish to look at page 3 of this article for color coding (kamma is the green one there):
http://www.biomedcentral.com/content/pdf/1471-2156-7-28.pdf
Coming to stories which probably may never be authenticated, kammas have generally held that a group of brahmins came off as an off-shoot from them. Possibly true, since andhra regions in buddhist times did have many preachers and teachers whose descendents might have found their way into the andhra brahmin population groups. i may wish to consider that they came down from the active buddhist beds such as bihar (which again explains the regional input).
Generally, it is possible to find individuals from all 3 - reddy, naidu, kamma groups clustering with speakers from kannada and marathi groups; and at times even sindhi speaking ones. However, there is no historical proof to show any old ancient migration from Pakistan-Sindh areas into Andhra. But it maybe likely such a migration did happen (probably after spreading across Gujarat, Madhya Pradesh, Bihar, Orissa, etc). Not strangely, all of these groups (telugu, kannada, marathi and sindhi) follow the lunar calendar, and minus the sindhi, quite a few customs b/w the telugu, kannada and marathi folk are common as well. In that way, I may want to consider that possibly apart from the ‘sani’ concept of an unofficial wife, the naming pattern for kammas may perhaps be related to the sindhi “ani” concept. But only a historian can offer any explanation or clarity on that.