If you look more closely at the species mixing disasters, the vast majority of the incidents involve torrent newts [Pachytriton] of one kind or another, species which are notoriously territorial even with their own kind, quite predatory on other amphibians, and constrained to habitats which are not suitable for most other amphibians anyway.
There are many possible species to mix, but I would always look at habitat, size,diet, and environment first. The lizard above,for instance, would not be suitable for several reasons: it's omnivorous, it's much bigger than the salamanders, and it occurs at lower altitudes, mostly lower latitudes, and warmer climates than the salamanders [although I have seen them survive freezing temperatures outdoors]. Despite toxicity of the salamanders, the lizard could kill them before the poisons make a difference, while the poisons in turn could harm the lizard.
If you want a more authentic mix, try searching for RAPD surveys for a region in which one or more of your target species occurs. Using such survey results will not only give you a more accurate assessment of potentially climatically compatible species, but will give more weight to your assignment results. Those surveys tend to cover all taxa for a region, typically something like a regional park or reserve, and climate data are generally available if not included. If you're lucky, you may also get habitat photos. From the plant surveys [you can also try searching for botanical surveys for particular regions [eg. google: botanical survey yunnan]], you can probably find plenty of substitute species for the rare or local endemics, as there are many groups with wide distributions. Viola for instance has species worldwide, while Pseudotsuga has species in China and the Pacific Northwest. Remember to research species groups and growth requirements though, as commonly available species may not look like or grow like the species surveyed. Australian Viola for instance, may grow as vines, while South American species often form tight rosettes. Rhododendron and Vaccinium may form trees in Asia, while they are shrubs or bushes in northern North America and Vaccinium are epiphytes in the neotropics.
Inverts could include Caridina, local terrestrial isopods, Hyalella, Gammarus, or Asellus/Coecidota, Palaemonetes. If these are not native to the region, they likely at least have relatives or analogs. Contrary to many views, fishes are a good possibility too: there are many small minnows [rasboras, danios, etc], loaches, and gobies which would occur naturally in the same water bodies as some of these salamanders [although more often in mountain streams than ponds]. Cousins of Bufo bufo and Hyla arborea are sympatric with Tylototriton, as are some Hypselotriton, Bombina maxima, and Japalura. Some Hylarana, Odorrana, Rhacophorus, Ptychozoon, and Theloderma can be expected as well. Most of these are not likely to be direct predators on either each other or your target species, although the more species you include, the more factors you have to consider. The main things are: everyone needs to eat, regardless of whether the same species or different, so enough food must be provided for all; and if breeding occurs, fish may eat eggs or larvae...but so will the parents.
Kadoorie Farms has done a number of RAPD surveys in China. As I have reviewed ALL of them for salamander localities, I can tell you that the following surveys/locations all had Tylototriton or Echinotriton recorded: 3*,13, 15, 18, 24, 29, 38
I developed a competition system for non-profit societies to use in the creation of "shows" which catch more of the flavor of dog or cat shows, in which people get together for the sake of an interest in the hobby and NOT simply to buy and sell products. One aspect of the system is the terrarium component. I'm attaching the current standard judging form for terraria, AND the "biotope" form. The latter is used as an add-on bonus for those who wish to not only create a good looking terrarium, but have it as close to a natural habitat as they can manage [that is, day geckos and bromeliads are not acceptable for a biotope]. If your habitats can score highly on both sheets, you're probably close to your goal.
* - a very interesting note. Jianfengling, SW Hainan, has records of both Echinotriton andersoni and Tylototriton hainanensis. The former seems a probable error based on confusion with the latter in an earlier publication. Definite records of Echinotriton come only from the Ryukyu Islands, Taiwan, and Zhejiang. Hainan is distant from these but plausible. Other taxa with multiple species groups have invaded the island repeatedly, to produce multiple related but endemic species. The gecko Goniurosaurus for example, has three species on Hainan, representing two species groups which have colonized the island separately. Hainan and Taiwan are located such that both Yaotriton and Echinotriton could have reached either island naturally.
Good design there, also the end enclosure is it a glass front to allow people to see in to the more aquatic species?
