C / C++ / MFC

//A) declare the array dynamically and allocate space for the variables.
double vars[]; // this is also equivalent to:  double *vars;
vars = new double[3];
...
//B) reallocate the array, to say 5 variables
double newVars[];
newVars = double[5];
// copy the data from the existing array before you delete the old array.
memcpy(newVars, vars, sizeof(double) * 3); // three is the current number of vars in the array
delete[] vars;
// copy the newVars array into your old array, now you can comtinue to use your old pointer.
vars = newVars;

//There is a little bit of setup work here, but do not let this intimidate you.
//This goes in your header file.
#include <map>
using std::map;

// This is a function object that is declared to help sort the string names in your map.
struct ltstr
{
  bool operator()(const char* s1, const char* s2) const
  {
    return strcmp(s1, s2) < 0;
  }
};

typedef map<char*, double, ltstr> mapVars;

mapVars vars;
// I read in your question that creating a name for your variables is not an issue, so for 
// simplicity I am going to use a statically defined string.
// Insert a new variable.
vars.insert("variable1", double*);
// Lookup the new variable.
double value = vars["variable1"];
//Remove this variable.  when you erase a variable from a vector or a dynamic array, you will have
//gaps, this is transparently handled with a map.
vars.erase("variable1");

// Once again, this goes into your header file.
// Also use the same code as above for the map.
#include <vector>
using std::vector;

typedef vector<double> vecDouble;
typedef map<char*, vector<double>, ltstr> mapVecVars;
// Here is how to use your new dynamic 2 dymensional variable vectors.
// First create the new varaible, which will actually be a vector.
mapVecVars vars;
vecDouble newVar;
// Add new doubles values into your new var.
newVar.push_back(1.0);
newVar.push_back(3.14);
newVar.push_back(98.6);
// Add the new vector of variables to your map.
vars.insert("variable1", newVar);

// You can now access the variables in "vars" like this:
cout << vars["variable1"][0] << endl;
// this would print out "1.0"
cout << vars["variable1"][2] << endl;
// this would print out "98.6"

// You can access the data that way because it is like calling this:
// vector<double> vec = vars["varaible1"];
// cout << vec[0] << endl;