Have to at least mention the Indricotherium.
It’s the largest land mammal known to have existed, about 8 m long, 5 m tall at the shoulder, and around 30 T.
Yes dinos are a subset of reptiles. But so are birds, and chameleons and alligators and…. that’s the whole point. One very small subset of one group evolved to gigantic proportions. They are atypical as no other subgroup is. I am not in any way implying that there are many other choices of land vertebrates of which reptiles are only one. I am stating outright that there are many other choices of land vertebrates equally distinctive as the dinos.
Saying “we only have reptiles, mammals, and birds” at the same taxonomic level as the dinos is incorrect. We have a huge number of subgroups just as deserving of the same level of classification as the dinos. A hundred or so within the reptilia alone. As you say, dinos are only a tiny subgroup of the reptiles, yet they alone developed huge size.
Then we have the hundreds of divisions within the amphibians. The hundreds within the syanpsids. The syanapsids and amphibians both had representatives that reached sizes larger than humans, and yet not one reached sizes equivalent to the dinosaurs. The dinosaurs are distinct oddities that widely settled on size as a solution. Even allowing for a skew caused by the fossilisation process, dinosaurs are much larger on average than these groups.
Not to get into a debate, but this fails to account for the survival of mammoths on uninhabited offshore islands until just a couple of thousand years ago, while they vanished from the mainland and inhabited islands.
It also seems to fail to take into account that human induced environmental change is known to have killed off more species than direct hunting.
Perhaps it’s a result of the decline of life on Earth in general. Food chains may no longer be robust enough to support many large herbivores or predators. I read recently (in a Scientific American article, I believe) that biological diversity reached its maximum about 100 million years ago and will gradually decline until only very simple organisms will be able to survive about 500 million years from now.
The reason for this is that, as the luminosity of the Sun increases, oxygen levels are gradually declining. In the past the level of oxygen was kept within narrow bounds by certain feedback mechanisms, but these become less efficient when sunlight gets too intense.
Sorry I can’t remember the details.
A bit of perspective is required about the dinosaurs, too.
The Diplodocus was 30 m long, but only weighed 10 T.
The Apatosaurus was 20 m long, and weighed 30 T, i.e., the same as the Indricotherium.
Only the Brachiosaurus truly dwarfed any land mammal. It was about 23 m long and weighed about 80 T.
For comparison, the blue whale can be up to 32 m long and 140 T.
T-Rex was late Cretaceous, Brachiosaurus was middle to late Jurassic.
Desmostylus I know that there have been a few big mammals. But there seem to have been more big dinosaurs. Indricotherium was immense, and a bit of an odity itself (and the weights were still open to debate last I heard, some putting the wieght at only about 2/3 that). There seem to have been quite a few dinsosaur species that topped the 20 tonne mark set by elephants.
It would be intersting to see the actual numbers of +20 tonne dinosaurs alive at any given time compared to any other group.It would also be interesting to see the average sizes and the numbers in the various quartiles.
Some dating estimates have Brachiosaurus becoming extinct during the Early Cretaceous, perhaps 130 mya; this gives 14 my as a Cretaceous genus, longer than genus Homo. The boundary between the Jurasssic and the Cretaceous was obviously less of a total extinction event than that between the Cretaceous and the Tertiary.
I would tend to think that the reason for gigantism might be found in the atmospheric composition;
it is possible that the giant insects and arthropods of the Carboniferous were able to exceed the sizes possible today because of a higher oxygen level in the atmosphere; similarly during the Mesozoic the carbon dioxide level is thought to have been high, and plant life may have beem prolific as a result.
Land animals might have evolved to take advantage of this abundance… although it is difficult to make an estimate of the population of each species after all this time.
That’s the real reason fo rejecting Kyoto. Huge grain harvests, and huge cattle to feed them to. 
The problem with attributing the die-off of large mammals to the end of the ice age is that, in the past 2.5 million years or so, there’ve been some 20 ice ages, each with a changing environment at its end. And only the most recent had an extinction of most large animals in North America.
So you have to ask what factor was different about the end of the most recent ice age that wasn’t present in previous ones. The only answer anyone has come up with is the arrival of humans. Now humans may not have hunted the mammoths and mastodons to extinction, but something they did was almost certainly the cause of their demise.
Damn, that thing looks just like a Walker from Star Wars.
It does, doesn’t it. 
Jasonfin,
That would be Carbon Dioxide levels, not Oxygen.
I have also read several articles and one book that put out the theory that the Earth as a habitable planet is in decline. 300-100 million years ago, Earth was robust as a habitable planet and now we live in a more impoverished world that will slowly grow worse. The trend will be for multi-celled animals will grow simpler and less complex as time goes on.
Time to crank up the space program!
Seriously… andtmurph, can you give references for the long-term decline? I’ve seen plenty of stuff that points at a local human-related decline, but it all assumes that the underlying baseline conditions on 100-million-year timescales are relaticvely constant.
Another animal worth mentioning in the “big and recently extinct” category is Steller’s sea cow.
About 10 m long, 8 T in weight, making it the biggest recent non-Cetacean mammal.
Discovered 1741, hunted to death by 1768.
Sunspace.
I can’t remember where I read the articles though Scientific American sounds familiar.
The book I believe (top of head) was ‘Life and Death of Planet Earth’ by Brownlee and Ward (?)
As an ecologist, I’ve always thought that there must have been something fundamentally different about the ecology of the Mesozoic to account for the extraordinary size of many dinosaurs.
What we need to consider here is the average size of the largest animals in communities. There is no question that the largest herbivores and carnivores in Jurassic and Cretaceous terrestrial communities - the dinosaurs - were far larger on average than those in any other communities before or since.
So what determines the largest size an animal can evolve to? It’s determined, at least in part, by amount of food that’s available to it within a given area - and that ultimately depends on the productivity of plant communities.
According to this site, carbon dioxide levels were higher - other sites say 10 times higher - during the Jurassic and in fact during much of the Mesozoic. (Some sites agree that temperature was also higher in the Jurassic, others dispute this).
Elevated carbon dioxide and elevated temperatures are both factors that can increase plant productivity.
My own pet theory is that dinosaurs were able to get so big because herbivores were able to harvest much more plant matter per unit time in a given area due to increased productivity from elevated CO2 levels. Carnivores got big because there was more food available for them as well, and they also had to be big to tackle the herbivores.
I thought this was a very interesting point you brought up, Blake. Rather than hijack this thread, though, I started a new thread on the topic here:
http://boards.straightdope.com/sdmb/showthread.php?s=&threadid=190709