The odyssey our genetic information undergoes as described in the Central Dogma of Molecular Biology is rife with potential pitfalls and yet we remain relatively - remarkably - unscathed. To shuttle our genetic information, messenger RNA (mRNA), must undergo modifications such as splicing. Splicing, which involves a plethora of small nuclear RNAs (snRNAs) and proteins, serves to remove introns and ligate flanking exons. The complexity of mRNA sequence recognition and spliceosome formation leaves the process vulnerable to mistakes such as the recognition of an erroneous site, hereafter referred to as cryptic splicing. In this thesis, I sought to elucidate the prevalence and peculiarities of cryptic splice sites to better understand the molecular characteristics governing their usage. To this end, I isolated, sequenced, categorized, and characterized these cryptic splice sites and discovered a number of sequence motifs and gene features that influence the erroneous usage of these sites. Subdividing the identified cryptic sites into 3 classes (recursive, high-fidelity, and low-fidelity), I assessed the predicted consequences of splicing at those sites with an explicit interest in sites that induce inclusion of a novel, cryptic exon. Additionally, I explored the therapeutic potential of redirecting cryptic splicing through steric inhibition of erroneous sites and subsequent skipping of cryptic exons. Ultimately, this thesis significantly expands our previously limited knowledge of cryptic splicing phenomena while better defining which classes of cryptic splice sites possess the greatest therapeutic potential.