A primitive word is that which can be traced no further to any English root; thus _circumspect, circumvent, circumstance, delude, concave_, and _complicate_, though compounds in the Latin, are to us primitives.
Derivatives, are all those that can be referred to any word in English of greater simplicity.
The derivatives I have referred to their primitives, with an accuracy sometimes needless; for who does not see that _remoteness_ comes from _remote_, _lovely_ from _love_, _concavity_ from _concave_, and _demonstrative_ from _demonstrate_?
But this grammatical exuberance the scheme of my work did not allow me to repress.

It is of great importance, in examining the general fabric of a language, to trace one word from another, by noting the usual modes of derivation and inflection; and uniformity must be preserved in systematical works; though sometimes at the expense of particular propriety.
Among other derivatives I have been careful to insert and elucidate the anomalous plurals of nouns and preterites of verbs, which in the Teutonic dialects are very frequent, and, though familiar to those who have always used them, interrupt and embarrass the learners of our language.
The two languages from which our primitives have been derived, are the Roman and Teutonic: under the Roman, I comprehend the French and provincial tongues; and under the Teutonic, range the Saxon, German, and all their kindred dialects.

Most of our polysyllables are Roman, and our words of one syllable are very often Teutonic.
In assigning the Roman original, it has perhaps sometimes happened that I have mentioned only the Latin, when the word was borrowed from the French; and considering myself as employed only in the illustration of my own language, I have not been very careful to observe whether the Latin would be pure or barbarous, or the French elegant or obsolete.
For the Teutonic etymologies, I am commonly indebted to Junius and Skinner, the only names which I have forborne to quote when I copied their books; not that I might appropriate their labors or usurp their honors, but that I might spare perpetual repetition by one general acknowledgment.

Of these, whom I ought not to mention but with the reverence due to instructors and benefactors, Junius appears to have excelled in extent of learning, and Skinner in rectitude of understanding.

Junius was accurately skilled in all the northern languages, Skinner probably examined the ancient and remoter dialects only by occasional inspection into dictionaries; but the learning of Junius is often of no other use than to show him a track by which he may deviate from his purpose, to which Skinner always presses forward by the shortest way.