diff --git a/docs/Data_Standardization/ontology.md b/docs/Data_Standardization/ontology.md
index ce27323..e3079b9 100644
--- a/docs/Data_Standardization/ontology.md
+++ b/docs/Data_Standardization/ontology.md
@@ -23,7 +23,7 @@ Most ontologies were created after the web itself, and as inter-agency collabora
 ## Advantages of ontology format
 Our data standardization writeup has explained the need for kinds of structured vocabulary covering agriculture and related domains with features that make them reusable, infrastructure friendly, and semantically precise.  Ontologies, combined with some curational best-practices, have these features built-in in a way that other structured vocabulary formats can't match.  Not all ontologies are created equal, and there are dead ones, and incomplete or very poorly designed ones.  Consequently, ontology collaboratives such as the OBO Foundry have created [curational principles](https://obofoundry.org/principles/fp-000-summary.html) to enable recognition of gold standard ontologies. We will also mention some other structured vocabulary formats like the common Simple Knowledge Organization System [SKOS](https://en.wikipedia.org/wiki/Simple_Knowledge_Organization_System) that though logically and/or semantically lax, may be useful since they have institutionally-backed resources, and meet the needs of library science style category indexes, like the [AGROVOC](https://www.fao.org/agrovoc/) agricultural concepts, definitions and relationships vocabulary.
 
-A brief note about ontology terminology below: One may open an OWL ontology in a popular ontology editor like Stanford's [Protege](https://protege.stanford.edu/) and see hierarchies and lists of terms in different places; mousing over a term will yeild a unique purl style identifier for it.  Ontologies express a few kinds of term: **classes** which are categories of things, **instances** which are things that (by explicit statement, or by reasoned inference) belong to one or more class categories, object properties that connect between classes or instances, and data properties that connect between instances (and sometimes classes) and particular values.
+A brief note about ontology terminology below: One may open an OWL ontology in a popular ontology editor like Stanford's [Protege](https://protege.stanford.edu/) and see hierarchies and lists of terms in different places; mousing over a term will yeild a unique purl style identifier for it.  Ontologies express a few kinds of term: **classes** which are categories of things, **instances** which are things that (by explicit statement, or by reasoned inference) belong to one or more class categories.  Two kinds of relation are offered, namely "object properties" like "part of" that connect between classes or instances, and "data properties" like "has value" that connect between instances (and sometimes classes) and particular values.
 
 A good ontology should be: