Discover
In the 1630s, the French lawyer Pierre de Fermat sent the Paris mathematicians his Method for determining maxima and minima and tangents to curved lines. Why would tangents and maxima belong in one method?
- At a maximum, the tangent is horizontal
- Both were needed to measure land
Key idea
One idea, found twice and written two ways.
- Step 1Pierre de Fermat c. 1607 – 1665Rolland Lefebvre. Public domain1630s: Fermat finds maxima, minima and tangents with one method and sends it to Paris.
- Step 2Isaac Newton 1642–1727Godfrey Kneller, 1689. Public domain1665: the plague closes Cambridge. At home in Lincolnshire, Newton develops his method of fluxions.
- Step 3Gottfried Wilhelm Leibniz 1646–1716Christoph Bernhard Francke, before 1729. Public domain1684: Leibniz publishes 'A new method for maxima, minima and tangents', with the d-notation we still use.
Newton spoke of fluxions, quantities that flow in time. Leibniz wrote and , and by 1676 he had the rule . We still write , his notation.
Nova methodus pro maximis et minimis, itemque tangentibus, quae nec fractas nec irrationales quantitates moratur, et singulare pro illis calculi genus.
A new method for maxima and minima, and also for tangents, which neither fractional nor irrational quantities hold up, and a singular kind of calculus for them.
What is Leibniz's 'new method' for?
- Finding maxima, minima and tangents
- Finding the areas of curved shapes
Worked example
Fermat's question, our way: which number makes as large as possible?
Practice ★
- In 1684 Leibniz published 'A new method for maxima, minima and tangents'. What did his paper give the world?
- His differential calculus, with dx and dy
- The first table of logarithms
- Newton's laws of motion
- With Leibniz's rule : find the slope of at .
- With Leibniz's rule : find the slope of at .
- A Fermat-style question: a rectangle has perimeter 108 cm. Which side length gives the largest area?
Summary card
Derivatives today
- Speed and acceleration in physics: rates of change.
- Economics: marginal cost is the derivative of the cost.
- Engineering and computer science: finding the best design, or training a neural network, means following slopes downhill.
Sources: MacTutor biographies of Fermat, Newton and Leibniz (see "Sources and credits").