>>OK, with Naomi's suggestion I realized I had the wrong field name in the select. So I fixed that and turned the relation around so that test_phases drives the result. This produces a 1 to many rather than a many to 1 result. I also made some changes to the oCA tables, the oCA.fieldlist, and the oCA.updatenamelist so that the field in test_phases would be updateable. Here is my code now and it seems to work at least to the point of reading and updateing both tables.
>>

>>this.oCa = createobject([CursorAdapter])
>>
>>text TO this.oCa.SelectCmd PRETEXT 15 NOSHOW
>>     SELECT test_financing.keyID, Test_financing.tip_id, Test_financing.amount,
>>            Test_financing.source, Test_financing._year, test_phases.phasename,
>>            test_phases.keyID as key2
>>            FROM \\tpdfiles\data\dfreeman\tip\data\TEST_phases.DBF
>>            left join test_financing on test_phases.tip_id = test_financing.tip_id and test_phases._year = test_financing._year
>>            WHERE  Test_phases._year = ( ?m.lnYear )
>>		    ORDER BY Test_phases.tip_id
>>endtext
>>
>>this.oCa.allowinsert		= .t.
>>this.oCa.insertcmd			= ""
>>this.oCa.buffermodeoverride = 3				&& 3 = row buffering, 5 = table buffering
>>this.oCa.datasourcetype = [NATIVE]			&& native fox datasource
>>this.oCa.sendupdates    = .t.				&& unnecessary except with ODBC or ADO
>>this.oCa.wheretype      = 1					&&
>>this.oCa.tables         = [test_phases,test_financing] 	&&
>>this.oCa.fetchsize      = -1				&& primarily for remote views
>>this.oCa.keyfieldlist   = [keyid, key2]			&& the primary key for the table
>>this.oCa.updatenamelist = 	"keyid test_financing.keyid," + ;
>>							"tip_id test_financing.tip_id," + ;
>>							"amount test_financing.Amount," + ;
>>							"source test_financing.Source," + ;
>>							"_year  test_financing._year," + ;
>>							"key2 test_phases.keyID," + ;
>>							"phasename test_phases.phasename"						
>>this.oCa.updatablefieldlist = "keyid, tip_id, amount, source, _year, phasename, key2 "
>>this.oCa.alias              = this.sourcetable
>>
>>m.lnYear = this._year
>>this.oCa.cursorfill()
>>

>>My next step will be to write code to provide INSERT and DELETE capabilities. What I am not sure of is this:
>>1. If the only fields that need to be updatable (except for INSERT and DELETE) are amount, source, and phasename, do the other fields need to be included in the updatenamelist?
>
>No, but you mus include KEY fields also in that list.
>
>>
>>2. Do I need to include the keys from both tables in the keyfieldlist?
>No, only the table you will UPDATE(insert, delete)
>
>
>>3. Do I need to include both tables in the tables list?
>Yes.
>
>
>>
>>4. What about the SELECT. Do I need to include more than the 2 keyfields and the fields that will be displayed?
>
>If you have to, yes.
>
>Keep in mind that CA can update ONLY one table unless you have custom command.

Well now I have received a curve ball. What would a custom command look like?
It seems the CA will update either table but only one or the other. If I edit them one at a time it appears to work, but if I edit them both at the same time it only saves one of the tables. Does that make sense? Also, if I edit one and then go to edit the other it looks like the first reverts to the original value. So maybe the CA isn't the best tool here?

Naomi - because I never heard of it before. This is my first attempt at using a CA and the bulk of the code (at least the outline) you see was provided by one of the helpers here at UT. (May have been Borislav or perhaps Cetin.) This is a project I have been playing with off and on for a couple of years now.