For example, i would like to query all reptiles of my lifeline which are poisonous or all plants which are carnivorous.
Setting up extra observation fields for that purpose is a bad idea because those are properties of the species and not of the observation.
Therefore there must be a way for querying such lists like mentioned above, but i have no idea how to accomplish it.
Not sure why this is a bad idea; if all individuals of a plant species are carnivorous, then the species and observation are the same thing re the observation field.
Because you need to enter the properties on every observation instead only one time at the species.
Besides, i am sometimes not sure which characteristics the organism has and have to check each time.
If you go to this link https://www.inaturalist.org/observations/thebeachcomber (but replace âthebeachcomberâ with your username), thereâs a button at the top which says âbatch editâ. Click this, then tick all the observations you want to edit. You can then apply a tag/annotation/make other changes that will be applied to all of them simultaneously.
Thanks, Iâve learned something new again 
But this does not answer my question.
It is wrong and illogical to edit the properties of the observations again and again, because they are always the same and does not change. On the other hand, the error rate is high.
It is correct and logical to define the properties of the species only once in the master data.
For example, Atropa Belladonna is a poisonous plant, that means the observation has to inherit the property âpoisonous plantâ of the master data.
iNat is about the observation, less so about the properties/characteristics of the taxa. There is information in the taxa page often derived from the likes of wikipedia, which may or may not have such information about the taxa, but to query the system to find âwhich taxa are poisonousâ for example, you are looking in the wrong place. To leverage iNat to accomplish that sort of task, you unfortunately do have to go through and apply the fields yourself. You could encourage others to help by making it into a project, eg âPoisonous things worldwideâ or perhaps leverage specialist knowledge by looking at projects on specific areas of taxa, such as âPoisonous amphibiansâ, or perhaps even geographically, and then others with similar interest would help in finding and including appropriate observations.
As an example, similar to your own Atropa example, iNat have explicitly stated that whether something is edible or poisonous is not something that iNat will determine in itâs taxonomic identifications. If you take peanuts for instance, some people are highly allergic and others not, so it is impossible and irresponsible to attempt to advise or determine such things in this platform
https://forum.inaturalist.org/t/create-a-foragable-or-edible-plants-search-option/6089
The above topic has some of the issues identified and discussed, and Ken-ichi comments from an âofficialâ position. He does raise the notion of crowd sourced annotation, but as you yourself point out, huge personal effort!
As several other responses have hinted, iNaturalist does not have âbuilt-inâ properties of species stored in its database. The taxonomy, observations, geography, etc., are all crowd-sourced in the iNat model, with some help from world-wide open-access databases for some of the taxonomy and geography.
So like everything else, we would have to find a way to âteachâ iNaturalist which plant taxa are carnivorous, for example, on a world-wide basis. Maybe such a database does exist, and could be attached to the iNat taxonomy, but that takes up limited staff time, and then what other taxon properties would we prioritize for such work?
So as others have hinted, we are left with the less-elegant solution of tracking properties at the observation level. The filtering capabilities on the web site make it pretty easy to find all observations of selected taxa at any level, for doing things like adding fields or tags, or collecting in a collection project. But in the end itâs a self-service model for such endeavors.
The simplest way right now is to create a collection project for the species in question. This may be really easy, or a huge pain depending on how many taxa fit into the category in question. For example, I just created this project showing all carnivorous plants in the world, based on the list on Wikipedia. Took me about five minutes to add all the taxa.
https://www.inaturalist.org/projects/carnivorous-plants-worldwide
Good idea, thank you 
But your project âcarnivorous plants worldwideâ does not list captive ones as well, e.g.:
https://www.inaturalist.org/observations/28352267
Why donât provide an interface for the community to edit the properties of the species? Like similar to the interface of observations, but for species only. Means for example that are at least 3 identical identifications necessary to be added to the main data of the species.
You might consider submitting a Feature Request topic to the forum, asking to allow filterable taxon-level âfieldsâ to be created by the community, analogous to the observation-level fields that currently exist. I canât promise that staff would decide to implement the request, but itâs worth a try.
Observation fields currently donât involve a âvoteâ - any user can add a category, define values, and assign a value to an observation. Iâm guessing there would be less appetite for a more complicated system analogous to Identifications, where community agreement would need to be tracked.
Speaking just for myself, I could see a taxon-level âLife Formâ field being very handy for plants, with multiple values allowed (annual, perennial, tree, shrub, carnivore, succulent, vine, parasite, epiphyte, etc.).
Iâve gone ahead and posted such a proposal here:
https://forum.inaturalist.org/t/create-taxon-field-functionality-analogous-to-observation-fields/7013