Summary: Intelligence is specific to a particular environment. A particular environment is populated by its own unique objects, and these objects are used to assess intelligence. As environments are particular to a species it follows that there is no pan-species scale of intelligence and that, consequently, intelligence is not manifested by any species but only by individuals within a species that share those objects.
Intelligence is assessed by assessing outcomes. Outcomes are profiled within a particular environment that supports skill sets particular to that environment. Intelligence describes a person's capacity to work or fit in with the environment associated with his/her species. Thus, a human is no more or less intelligent than an amoeba: each has its own unique environment or set of objects and a corresponding unique set of skills.
Intelligence is not manifested by any species or any type of object, whether animate or inanimate. This does not preclude the manifestation or measurement of intelligence between individuals within a species. This proposal configures intelligence as transcendentally ideal; that is, the objects of intelligence (species-specific individuals) are manifested through a framework. In this case the framework is a species. A species supports a unique set of objects that are enjoyed by its individuals. This set is commonly (and, invariably, anthropomorphically) referred to as the species' environment. It follows that without that framework intelligence cannot be manifested or measured.
Intelligence manifests between individuals that share an environment or set of world objects. Environments are not sets of geographical points but are unique to a species. Species share a geography, but species-specific individuals share an environment. For example, bacteria and I share the geographical kitchen, but our kitchen environments are different. Not only are our environments different, but they also contain different objects, objects that cannot be recognised by the other species (hence Wittgenstein's reminder that if a lion could speak we could not understand it).
Intelligence tests present us with objects that are specific to the human environment, simple objects such as signs or letters and numbers, and the cultural habits and mores that configure them as object-outcomes. This capacity that establishes IQ and establishes how we fit into our environment is itself a culturally-contingent, value-laden judgement, and because of that, IQ has no re-identifiable points on its scale - it is not absolute, and there are no objects, either animate, inanimate, species, or particular individuals, that can be reidentified by means of it.
Intelligence may be associated with consciousness but there is no epistemological gain to be made from making, or denying, the link. It is all the same to establishing a human-individual or amoeba-individual scale of intelligence whether or how the human or ameoba is conscious or not. In its business of being for itself the amoeba's "inner" life might be a blaze of light: it would not matter, not least because it could be argued that in the absence of common amoeba-human objects we could not, on principle, know of such light - and it would not be any of our business if, like a god, we could know.