In this thesis I present new data on long non-coding RNAs (lncRNAs) from one of the earliest diverging lineages in the animal kingdom. The ultimate motivation for this work was to understand how unicellular eukaryotes evolved into complex multicellular animals, and how regulatory RNAs are involved in gene regulation and development of the body axis in ancestral animals. The primary study organism is Sycon ciliatum, a calcareous sponge living in marine coastal areas. Sponges have a very simple morphology and belong to one of the deepest branches in the animal phylogeny. They are therefore useful for reconstruction of the last common animal ancestor. The main questions I address in this thesis are; does S. ciliatum possess and express lncRNAs and are these differentially and specifically expressed along the primary axis? Are there any similarities in the expression of lncRNAs and coding genes known to be involved in axial formation, such as Wnt and carbonic anhydrase genes? Do lncRNAs have a global expression pattern that resembles the global expression pattern of coding of genes? And, are lncRNAs of Sycon ciliatum conserved among other basal animals? I have addressed these questions by multiple complementary approaches to reveal the spatial and temporal dynamics of the lncRNAs in this species. In situ hybridization has been used to reveal cell specific expression of lncRNAs upregulated in the top part of the sponge. Bioinformatic approaches have been used to identify coding genes and lncRNAs that are co-expressed as the sponge regenerates its primary body axis, as well as differentially expressed lncRNAs between the regenerating samples. The genes co-expressed with lncRNAs have been functionally annotated in order to try to assign functional categories to these novel lncRNA genes. Results from these experiments clearly show cell-specific expression of lncRNA genes, and some even resemble the spatial expression of known genes involved in axis formation such as Wnt genes and genes involved in development of spicule-forming cells. In the final section I present preliminary results on lncRNAs generated from another deep animal lineage, the comb jellies, recently proposed as the very earliest branch in the animal phylogeny.