R help - Aug 2009 - Fwd: Overlay on raster loses points when merging with data frame

Does anyone have an idea about this, it's driving me insane!

Hello everyone.  I have overlayed 1611 points on top of a tif file.

tifclassoverlay<-overlay(CRS11, classifiedmodrobin)

overlaydf<-as.data.frame(tifclassoverlay)

tifol<-remove.na.rows(overlaydf)


tifol is 854 rows long.   tifol has predicted data, while the
classifiedmodrobin matrix has observed data.  When I try to merge the two, I
get a length of 827:

 ngafulldf<-as.data.frame(classifiedmodrobin)

 NGAfiledata<-merge(tifol, ngafulldf,all=FALSE, by=c("LON",
"LAT"))

NGAfiledata<-unique(NGAfiledata)

NGAfiledata<-remove.na.rows(NGAfiledata)
The length should obviously still be 854.  What happened to the missing
rows?  How can I get them back?  Thank you very much!! Here is a sample of
what the data looks like:
        LON     LAT Net  STA ELEVATION vs30 tif.cat
1  -3086165 3718184  CI NEE2       271  363       8
2  -3143337 3500449  CI  EMS        11  336      14
3  -3155773 3704130  CI  DAN       428  304       4
4  -3173788 3722206  CI  GRP       978  509       4
5  -3188256 3548775  EN  SIM       -64  264      NA
6  -3193879 3549663  EN  RED       -64  322      NA
7  -3195312 3547901  CI  RXH       -59  331      NA
8  -3214063 3540800  CI  ERR       -56  233       8
9  -3219756 3587780  CI NSS2        17  246       4
10 -3226717 3558278  CI  SAL        14  473       8
11 -3229979 3516218  CI  CRR        66  315       4
12 -3238308 3596300  CI  THX       -31  244      12
13 -3247148 3667171  CI  GTM       836  306       7
14 -3254339 3615978  CI  MGE        68  330       4
15 -3257283 3597678  CI  PMD       296  257       3

I am assuming there is no precision error since there are no decimals..

Thanks!

Mehdi Khan

	[[alternative HTML version deleted]]