Gnuplot, superimpose curves with splot - gnuplot

Question

I am trying to plot a family of curves in space using gnuplot.
I am trying to do it with for-loops. The data are in 4 columns, and every 20 rows, a new plot starts, e.g., if I do a plot using the 3 and 4 columns, on the row 21, the index in the third column starts again, and the number on the second column increases by one. This pattern repeats, and every 320 rows, the index on the first column increases by one, and all the pattern repeats itself again.
The index in the first column goes from 2.0 to 3.8 whereas the index on the second column goes from 75 to 90.
In my code I tried to set the range limits from 1.8 to 4.2 for one axis, and from 0 to 100 for the second axis.
I wrote the following code
a=0 # this variable indicates how many rows to skip
do for [j=20:38:1]{ # loop for the first column
do for [i=75:90:1]{ # loop for the second column #
splot [1.8:4.2][1:101] "dataforgnuplot" every ::a::a+19 u 1:3:4
w lines title sprintf("%3.2f %3.2f",j/10.0, i);
a=a+20;
}
}
and it generates one curve at a time, but the previous one disappears, I had a look at multiplot, and also was trying with replot, but, I still haven't found how to do that.

Answer 1

You should approach this differently. If you put blank lines in your input file in the right places, and plot the whole file, gnuplot will know exactly when to stop drawing one curve and start the next one.
From the gnuplot documentation:
Single blank records designate discontinuities in a plot; no line will join points separated by a blank records (if they are plotted with a line style).
Two blank records in a row indicate a break between separate data sets. See index (p. 79).
The difference between the two cases is that separate curves which are part of the same data set (one blank line) will be plotted in the same color and with a single entry in the legend. Different data sets will have different colors and separate labels.
It should be easy to change the program which outputs this data if you have control over it, to add the extra blank lines. If you don't control it, a one-or-two-line script should be able to insert the blank lines, either based on the line counts you gave, or by checking the first and second data records.