Abstract 353 16.1 Introduction 354 16.2 Steady-State Systems 354 16.3 Structural Inheritance 355

16.3.1 Inheritance of Cell Surface Structure 355 16.3.2 Genetic Membranes 356 16.3.3 Prions 356

16.4 Inheritance of Chromatin Marks: Some Mechanistical Considerations 357 16.4.1 DNA Methylation 357 16.4.2 Chromatin Remodeling 358 16.4.3 Histone Acetylation 358 16.4.4 RNA-Mediated Gene Silencing 359 16.4.5 Fidelity of Transmission and Epimutations 359

16.5 Inheritance of Chromatin Marks through Meiosis 359 16.5.1 Is Epigenetic Inheritance a Specific Property of Certain

Genomic Regions? 361 16.6 Evolutionary Potential 362

16.6.1 Why k<1/2 Might Often Be the Case 362 16.6.2 Why k > 1/2 Might Be Found 363 16.6.3 Range of Heritable Epigenetic Variation 364

16.7 Discussion 365 References 367

Although epigenetic inheritance has been recognized to be crucial to maintain different cellular states during development, it is still unclear whether and how often epigenetic marks can be important in adaptation. As epigenetic inheritance encapsulates a wide range of phenomena, we first briefly describe the mechanisms behind the heritable potential of (1) metabolic steady-state systems, (2) cellular structural elements and (3) chromatin marks including DNA methylation. Next, we discuss the experimental evidences for the transmission of chromatin marks

through meiosis. Although these results provide a clear mechanistic basis for heritable epigenetic variation, an important possible objection is that they might not be stably transmitted through meiosis (and hence not between generations). Moreover, in many cases, epigenetic marks are not inherited in a Mendelian fashion; they are either transmitted to too many progeny, in which case they can also be deleterious, or to too few, in which case even if advantageous they will often be lost. This suggests that under sexual reproduction — possibly associated with cell fusion — epigenetic inheritance is unlikely to play a fair Mendelian game, which is a prerequisite for adaptive evolution.