CSV Header Rows

Hello.

I am seeing some behaviour I don't understand. I can't believe it hasn't been discussed here before, but my searches have returned nothing.

Here it is.

I have a CSV file. Its first rows look like this:

# 3_results/ABC.csv
# Created from data/clean/
# OS Unix, host smac.local, user ge
col1,col2,col3,col4,col5,col6,col7,col8,col9,col10,col11,col12,col13
38, 2008-05-05 14:35:05, 92,0,0,FFF,1,0,1,0,1,0,0
41, 2008-05-05 14:35:10, 4,1,0,GGG,1,0,0,1,0,0,0
75, 2008-05-05 14:35:26, 2,1,0,HHH,1,0,0,1,0,0,0
183, 2008-05-05 14:35:26, 4,1,0,III,0,0,0,0,0,1,0
347, 2008-05-05 14:35:28, 11,0,0,JJJ,0,0,0,0,0,1,0
365, 2008-05-05 14:35:28, 0,1,0,KKK,0,0,0,1,0,0,0
41, 2008-05-05 14:35:29, 18,0,0,LLL,1,0,0,1,0,0,0
286, 2008-05-05 14:35:30, 3,1,0,MMM,0,0,0,0,1,0,0

So, there are three lines of miscellaneous information, one line of column names, and the data to be imported begins on row 5.

So I import with this:

BULK INSERT tableName FROM 'file.csv'
WITH (FIELDTERMINATOR = ',',FIRSTROW = 5)

And the first line to be imported is the one beginning with 183!

This does what I want:

BULK INSERT tableName FROM 'file.csv'
WITH (FIELDTERMINATOR = ',',FIRSTROW = 2)

I can insert as many lines starting with # as I want and it works fine.

What is going on?

Thanks.