[R] How do I tapply to a data frame with arbitrary column labels?

[Bill McNeill (UW)]

I have a data file that looks like this

class    pigeon    falcon    propeller    jet    wing    fly
birds    25           37          0                0      2        1
planes 0             1            28              40    1        3
birds    19           41          0                1      4        6
planes  0            0            25              50    5        5
planes  1            0           30               45    3        5

There may be an arbitrary number of unique class labels and an
arbitrary number of uniquely named columns.  I use read.table to read
my file into a data frame.  I then want to take the sum of all the
column values for each of the class types.  That is I want to end up
with:

class   pigeon    falcon    propeller    jet    wing    fly
birds    44          78          0                1      6         7
planes 1             1            83              135  9         13

where (pigeon, birds) = 25 + 19 = 44, (propeller, planes) = 28 + 25 +
30 = 83, etc.

I see how to use tapply to get the sum for each individual column, e.g.

> tapply(data$falcon, data$class, sum)
 birds planes
    78      1

But I can't figure out what command will do these sums for the entire
data frame without me having to specify the column names in the
command.

Thanks.
--
Bill McNeill
http://staff.washington.edu/billmcn/index.shtml