Also we know in the past there were dargonflys with wingspans greater than 2 feet and centipedes that grew as long as a car.
Nonetheless, the biggest dinosaurs were several times heavier than the heaviest terrestrial mammals that have ever lived, which in turn were several times heavier than the biggest elephants of today.
So probably dinosaurs were close to the size limit that is imposed by the mechanical properties of vertebrate tissues like bones, tendons and muscles, while in the present terrestrial mammals the size is limited by the availability of food and by the humans who have always hunted preferentially the biggest animals they encountered, and not by the strength of their bones.
This book fails to mention both Gideon Mantell (1790–1852) and Richard Owen (1804–92). It's through them that dinosaurs entered our modern, scientific world.
What IS true though is that pterosaurs and later birds and bats came onto the scene, and that's just game over for giant dragonflies.
An efficient circulatory system and myelinated nerves.
At small sizes, the circulatory and respiratory systems of insects work better (less friction for blood in the absence of capillaries and less friction for the air that circulates as a gas through the body), but at big sizes those of vertebrates work better, due to efficient pumping.
The myelinated nerves have the same speed for propagating nervous signals as non-myelinated nerves that are much thicker, so thick that an insect could not have many such nerves. At small sizes this does not matter, but at big sizes if the nerves are not fast enough the reaction times of an animal can become too slow, e.g. the head might need a big fraction of a second until noticing that something bites the tail.
The combined results are that an insect as big as a typical vertebrate would react mostly sluggishly and it would be able of only very short bursts of intense activity, before becoming tired. This would put it at a serious disadvantage.
It's also possible that for larger creatures a less dense and more flexible skeleton material (in general) might be ideal. Especially for animals that are not warm blooded (larger volume to mass ratio might resist temperature changes better?)