Decrypt multidimensional arrays allocated dynamically.

So, I have this code, and I am trying to deallocate the array ppint at the end. I have tried using leaks with Xcode to figure out if it is working, but I don't quite understand it. Will doing this work?

delete ppint[0];
delete ppint[1];
delete ppint[2];
delete ppint[3];

Or is there something else that must be done?

#include <iostream>
#include <string>
#include <unistd.h>
using namespace std;

int main()
{
    int **ppint;
    ppint = new int * [4];

    for(int i = 0; i < 4; i++ ) {
        ppint [i] = new int[4];
    } // declares second layer of arrays

    for(int i = 0, count = 0; i < 4; i++ ) {
        for(int j = 0; j < 4; j++ ) {
            count++;
            ppint [i] [j] = count;
        } //init part 2
    } // init array

    for(int i = 0; i < 4; i++ ) {
        for(int j = 0; j < 4; j++ ) {
            cout << ppint [i] [j] << endl;
        } // print part 2
    } //print array
}

Close. You would have to use delete[] because you allocated each of them with new[]:

delete[] ppint[0];
delete[] ppint[1];
delete[] ppint[2];
delete[] ppint[3];

But of course, you should use a loop:

for(int i = 0; i < 4; i++ ) {
  delete[] ppint[i];
}

And then don't forget to delete[] ppint itself:

delete[] ppint;

However, in C++, we prefer to not mess around with dynamically allocated arrays. Use a std::vector<std::vector<int>> or a std::array<std::array<int, 4>, 4>. If you care about locality of data, try boost::multi_array.