What is a limitation of using a ball-and-stick model for dna?
a) some kinds of atoms that are found in dna must be left out of the models. b) dna has so many atoms that the size and cost of a model could be too great. c) it is impossible to distinguish different kinds of atoms with models. d) the chemical bonds between the atoms in dna must be part of the models.

I would say B. My reasoning behind this is:

Most models can't incorporate all the details of complex natural phenomena. For example, when measuring distances around the Earth it's convenient to model the Earth as a sphere, but this doesn't incorporate variations in distance because of mountain ranges, valleys and other topological features the traveler must traverse. Incorporating these additional details would make the model too complex for easy use. Since models must be simple enough that you can use them to make predictions, they often leave out some of the details.