normalize a histogram in pdf mode

I am trying to normalize my histograms so that my gaussian curves can fit over them (right now, the histograms and gaussians are not on the same scale so the gaussians look really tiny. How do I fix this? I tried doing histogram(midterm,bins)./sum(histogram(midterm,bins)) but matlab didn't like that and said "Undefined function 'sum' for input arguments of type 'matlab.graphics.chart.primitive.Histogram'."

[num, txt, raw] = xlsread('grades.xlsx')
midterm = num(:,2);
t = tabulate(midterm);
scores = t(:,1);
counts = t(:,2);
assign = num(:,1);
bins = 0:15;
firstp = makeNormalDis(13.8306,1.3311);
secondp = makeNormalDis(10.7675,1.9410);
x = [0,15];

figure; 
a = histogram(midterm, bins); hold on;
b = histogram(assign, bins);
fplot(firstp, x, 'r');
set(findobj(gca, 'Type', 'Line', 'Color', 'r'), 'LineWidth', 1.5);
fplot(secondp, x, 'b');
set(findobj(gca, 'Type', 'Line', 'Color', 'b'), 'LineWidth', 1.5); hold off
set(gca,'XTick',bins);
ylabel('Normalized PDF');
xlabel('Grades');
legend('Midterm', 'Assignment 1', 'Location', 'northwest');
grid on;

Answer 1

[the cyclist]

Set the Normalization property of the histogram object. Here is an example:

h = histogram(randn(100000,1));
h.Normalization='probability'