how to add two data.frame with the same column but different row numbers

Thanks, Dennis! I'll go with it. It's surprising there is no ready way to do
that. I imagine it should be a common data manipulation to add two
data.frame from two different sources. It could happen that one data.frame
is missing some rows while the other have some more.

If you represent them as zoo series then you can do it using +
(although the definition of + is different than in your post).   Here
"a", "b" and "c" are the "times":

library(zoo)
a <- zoo(1:3, letters[1:3])
b <- zoo(c(6, 1), c("a", "c"))
a+b

The last line gives:

a+b

a c
7 4

To use the definition in your post one could do this (which has the
effect of modifying b so that a+b works as in your post):

merge(a, b, fill = 0, retclass = NULL)
a+b

The last line gives:

a+b

a b c
7 2 4

Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com