Many complexities in our world come about through the use of preexisting purposeful information. This information may be structured in various ways (e.g., instructions, recipes, algorithms, rules, rules of thumb, business plans, and expert knowledge) and, if followed, directs the formation of something which otherwise would not have existed. This chapter argues that information organized in this way must ultimately arise as the output of an evolutionary computation. Because of this, an evolutionary process underlies most everything that characterizes human existence. This principle includes economics and markets. This chapter addresses whether or not understanding the fundamental role of evolutionary computation for enabling human and biological complexity provides useful insight into market behaviors and introduces the basic concepts necessary to have this discussion.

Complex systems theory and evolutionary theory hold important insight for economics, yet to date they have played a limited role in shaping modern economic theory. This chapter reviews different notions of equilibrium and explores four distinct areas relevant to the incorporation of evolutionary and complexity ideas into economics, finance, and policy. It investigates the determinants of major economic transitions, such as the Industrial Revolution or the collapse of the Soviet Union. It asks whether evolutionary processes should lead to an increase in complexity, on average, of economic and social systems over time. It reviews modern theories of group learning in biology, which have both evolutionary and complexity dimensions, to see if they might be relevant to human social institutions, such as firms. It analyzes whether the structure of human interactions or individual human intelligence is primarily responsible for the performance of our institutions. Finally, it finds the methods of evolutionary analysis and of complex systems to be extremely useful in capturing the open-ended, evolving nature of an economy composed of interactive agents and suggests that these methods be used to create more realistic models of actual markets and economies.

The concepts of evolution and complexity theory have become part of the intellectual ether permeating the life sciences, the social and behavioral sciences, and, more recently, management science and economics. In this book, John E. Mayfield elegantly synthesizes core concepts from multiple disciplines to offer a new approach to understanding how evolution works and how complex organisms, structures, organizations, and social orders can and do arise based on information theory and computational science. Intended for the intellectually adventuresome, this book challenges and rewards readers with a nuanced understanding of evolution and complexity that offers consistent, durable, and coherent explanations for major aspects of our life experiences. Numerous examples throughout the book illustrate evolution and complexity formation in action and highlight the core function of computation lying at the work's heart.

In computational theory, time is defined in terms of steps, and steps are defined by the computational process. Because steps can be described, a computation can be recorded as a binary string. This allows time to be measured in bits, which in turn, allows the definition of various computable complexity measures that account for the minimal amount of computation required to create an object from primitive beginnings. Three such measures are introduced in this article. They are “transcript depth,” which is closely relate to logical and computational depth, and “ Kd complexity,” which is similar to Levin's Kt complexity, and “minimal history.” The later two measures are comprehensive in the sense that they characterize all information required for the creation of an object and also all computable internal relationships and redundancies that are present in the object. © 2007 Wiley Periodicals, Inc. Complexity 12:48–53, 2007 This paper was submitted as an invited paper resulting from the “Understanding Complex Systems” conference held at the University of Illinois–Urbana Champaign, May 2005

We investigate the following question. Do populations of evolving agents adapt only to their recent environment or do general adaptive features appear over time? We find statistically significant appearance of general adaptive features in a spatially distributed population of prisoner's dilemma playing agents in a noisy environment. Multiple populations are evolved in an evolutionary algorithm structured as a cellular automaton with states drawn from a rich set of prisoner's dilemma strategies. Populations are sampled early and at the end of a ten-thousand generation simulation. Modern and archaic populations are then placed in competition. We test the hypothesis that competition between an archaic and modern population yields probability p=0.5 of modern populations out-competing archaic ones. The hypothesis is rejected at a confidence level of 99.5% using a binomial probability model in each of seven variations of our basic experiment.

The organization of repetitive DNA sequences has been investigated in bovine DNA. Repetitive sequences of all kinds constitute 25% to 30% of the total. Five density satellites constitute about 20% of the genome, and most of the remainder consists of alternating repeating and nonrepeating sequences. The nonrepeating sequences have a very broad size distribution averaging 4,000 nucleotide pairs in length, with the longest exceeding 10,000 nucleotide pairs. The interspersed repetitive sequences are much more nearly homogeneous in size, averaging 350 nucleotide pairs in length, and are divided into 8 to 14 sequence families.

ABSTRACT We report the cloning and sequencing of the Brucella abortus oxyR homolog and provide evidence that the transcription product of this gene binds to the B. abortus catalase promoter region. A gene replacement/deletion Brucella oxyR mutant exhibits increased sensitivity to prolonged exposure to H 2 O 2 and is unable to adapt to H 2 O 2 in the environment.

Phylogenies based on nine prokaryotic catalase sequences demonstrate no relationship to phylogenies based on rDNA sequences or other known criteria. When this observation is considered together with the monophyletic relationship observed for eukaryotic catalase sequences, it seems likely that the catalase gene sequence has migrated repeatedly from eukaryotes to prokaryotes.

Mitochondrial DNA (mtDNA) was isolated from leukocytes contained in whole blood of cattle. Leukocyte membranes except the nuclear envelope were solubilized in a buffer that contained 1% Triton X-100. After sedimentation of cell nuclei, mtDNA was purified from the cell lysate by organic solvent extraction and ethanol precipitation. Approximately 5 µg of mtDNA was recovered from 400 ml of whole blood, a quantity sufficient for routine DNA cloning procedures or for detailed restriction mapping studies. mtDNA isolated with this method is a suitable substrate for several DNA-modifying enzymes. Thus, preparation of mtDNA from blood by detergent lysis provides a noninvasive alternative to tissue biopsy for characterization of mitochondrial genotypes in studies of evolutionary genetics and population dynamics.

Inheritance of mitochondrial DNA (mtDNA) in Holstein cattle was characterized by pedigree analysis of nucleotide sequence variation. mtDNA was purified from leukocytes of 174 individuals representing 35 independent maternal lineages, and analyzed for nucleotide sequence variation by characterization of restriction fragment length polymorphism and direct sequence determination. These data revealed 11 maternal lineages in which leukocytes from some individuals seemingly were homoplasmic for the reference mtDNA sequence at nucleotide 364, whereas those from other individuals were homoplasmic for a sequence variant at this position. Both alternative alleles were detected in all branches of these 11 lineages, suggesting that mutation at nucleotide 364 and fixation of the variant sequence occurred frequently in independent events. Thirteen instances were detected of mother-daughter pairs in which leukocytes of each of the two animals seemingly were homoplasmic for a different allele at nucleotide 364, demonstrating the bovine mitochondrial genome can be replaced completely by a nucleotide sequence variant within a single generation. The two alternative sequences seemingly arose de novo at similar frequency, ruling out replicative advantage or other selective bias as the explanation for rapid fixation of mutations at nucleotide 364. Another instance of intralineage sequence variation was detected at nucleotide 5602. This variation was detected in only one of the lineages examined, and evidently arose within three generations.

A major family of short, interspersed, repeated sequences in the bovine genome has been characterized. This family makes up the majority of all non-satellite repetitive DNA or about 6% of the bovine genome. It is estimated that there are at least 600 000 copies of this family interspersed among non-repetitive DNA sequences. Sequence analysis shows that this family includes sequences reported previously by Watanabe et al. (Nucleic Acids Res. 10, 1459-1469, 1982) and is distantly related to the human Alu sequence family.

Partially denatured DNAs from mouse, cow, and chicken were visualized in the electron microscope by the basic protein film technique and the size and distribution of the denatured regions characterized. A-T rich sequences visualized at 15% denaturation average about 1500 bases in length for all three species and are arranged quite non-randomly in the genome. This arrangement is such that 30–50% of the entire genome contains no A-T rich DNA, and another 20% is composed about one-half of A-T rich sequences and one-half of other sequences. Comparison with DNA denaturation profiles indicates that for each organism these sequences are from 25–35% G+C and that there is very little if any DNA more A-T rich than these. Estimates from published studies of fluorescence enhancement of quinacrine bound to A-T rich DNAs suggest that the observed non-random organization of A-T rich sequences is sufficient to account for Q banding of metaphase chromosomes.

Partially denatured DNAs from mouse, cow, and chicken were visualized in the electron microscope by the basic protein film technique and the size and distribution of the denatured regions characterized. A-T rich sequences visualized at 15% denaturation average about 1500 bases in length for all three species and are arranged quite non-randomly in the genome. This arrangement is such that 30--50% of the entire genome contains no A-T rich DNA, and another 20% is composed about one-half of A-T rich sequences and one-half of other sequences. Comparison with DNA denaturation profiles indicates that for each organism these sequences are from 25--30% G+C and that there is very little if any DNA more A-T rich than these. Estimates from published studies of fluorescence enhancement of quinacrine bound to A-T rich DNAs suggest that the observed non-random organization of A-T rich sequences is sufficient to account for Q banding of metaphase chromosomes.

Moderate resolution thermal denaturation profiles are presented for the purified DNAs from Escherichia coli, cow, mouse, rat and chicken. All show multiple thermal transitions indicative of large blocks of DNA with very similar base composition. The eucaryotes all have much more of this kind of DNA than does E. coli. The satellite DNAs of cow and mouse are clearly visible and it is likely that the other transitions represent additional families of repeated DNA.

Nuclei from various mouse tissues exhibit a pattern of fluorescence characteristic of the cell type when stained with the fluorescent compound Hoechst 33258. When such preparations are hybridized in situ with 3H-RNA complementary to the A-T rich satellite of mouse, it is clearly seen that only the fluorescent regions of the nuclei contain the satellite DNA. Thus Hoechst 33258 allows the precise localization of satellite DNA at all stages of the mouse cell cycle.

Cytological evidence is presented which shows that for Drosophila virilis and Samoaia leonensis at least, each satellite DNA is condensed into a distinct heterochromatic mass during interphase. This is seen as just one example of a general phenomenon in which chromatin containing a particular DNA sequence binds to other chromatin containing the same sequence. It is proposed that DNA sequence specific proteins can account for this phenomenon.

Chromosomal RNA was isolated from several rat tissues by a new technique. It is shown by RNA-DNA hybridization that each tissue contains a distinct population of chromosomal RNA sequences. This finding is compatible with the proposal that chromosomal RNA is involved in the process of gene activation. Chromosomal RNA (cRNA) is a class of nuclear RNA distinguished by its sequence diversity, small size, and high dihydropyrimidine content (1). Several experiments suggest that this RNA, apparently found in all eukaryotes, provides the specificity of gene control (2, 3). Assuming this to be true, there can be only two possibilities; cRNA is either a gene activator or a gene repressor. In a typical eukaryotic cell, 90% or more of the genes are repressed. Thus, when one cell type is compared with another, the active genes are usually different, but the repressed genes are largely shared. Any gene regulator must reflect the genes it controls. Thus, it follows that if cRNA is an activator, every tissue should contain a distinctive set of cRNA sequences. If, on the other hand, it is a repressor, the cRNA sequences should be much the same from tissue to tissue. The hybridization data presented in this paper clearly show that the cRNA sequences of the rat tissues investigated are quite different. This result is consistent with the hypothesis that cRNA is a gene activator. MATERIALS AND METHODS Tissues. All tissues were obtained from 200-400 g, male, albino rats of the Sprague-Dawley strain purchased from Berkeley Pacific Laboratories. Regenerating livers were produced by surgical removal of the two largest lobes of the rats' livers by the method of Higgins and Anderson (4). Novikoff ascites-tumor cells were obtained from rats injected 6 days earlier with 0.5 ml of ascites fluid. The tumor line was maintained by serial transfer of 0.5 ml of ascites fluid on a 6-day cycle. Chromatin. Crude chromatin was isolated from the solid tissues, as described by Marushige and Bonner (5), with three final washes in 0.01 M Tris HCl buffer (pH 8). Nuclei were made from the ascites tumor cells with 0.5%7O Triton X-100 in saline-EDTA buffer (0.75 M NaClI-.024 M EDTA, pH 8), and chromatin was isolated from the nuclei by three cycles of Abbreviations: DOC, deoxycholate; SSC, standard saline citrate (0.15 M NaCl-0.015 M Na citrate). * Address reprint requests to James Bonner. Reprints will not be available to residents of North America. 2652 homogenization in a glass-Teflon homogenizer, followed by centrifugation at 10,000 rpm for 10 min, all in 0.01 M TrisHCl buffer (pH 8). Preparation of Chromosomal RNA (cRNA). cRNA was prepared from crude chromatin in two ways. The first, termed the "CsCl method", was exactly as described by Dahmus and McConnell (6). Chromatin was made 4 M in CsCl by the addition of 6 M CsCl-0.1 M Tris * HCl (pH 8), and centrifuged for 12-20 hr at 36,000 rpm in the Spinco 40 rotor. Protein floats to the top to form a pellicle or "skin", and cRNA is obtained from this skin by Pronase digestion, phenol extraction, and elution from DEAE-Sephadex A-25 with a 0.2-1 M NaCl gradient, all in 7 M urea. cRNA elutes as a sharp, symmetrical peak at 0.55 M NaCl. This method works very well for ascites tumor, but poorly for rat liver, and not at all for rat kidney. A second method was, therefore, developed. This method, which will be referred to as the "DOC method", yields cRNA reliably from all tissues investigated. An equal volume of 1.7 M sucrose (enzyme grade, Mann Research) is added to the crude chromatin and the mixture is homogenized in a glass-Teflon homogenizer (3-5 strokes). Then, with constant stirring, 0.3 M sodium deoxycholate, is added dropwise, to a final concentration of 15 mMi and the mixture is stirred for an additional 30 min at 40C. It is next centrifuged at 36,000 rpm for 6-12 hr in the Spinco 40 rotor. The supernatant is decanted and made 7 M in urea and 0.2 M in NaCl by the addition of solid urea and NaCl. DEAESephadex, preequilibrated with 7 M urea-0.2M NaCl, is then stirred in and a 0.9 X 20 cm column (10-50 g of tissue) is poured with the slurry. When the column has been poured and all the liquid has passed through, the cRNA is eluted with a 0.2-1 M NaCl gradient in 7 M urea. As with the CsCl method, the cRNA comes off in a homogeneous peak at 0.55 M NaCl. The RNA is routinely eluted a second time from DEAESephadex to remove residual impurities. All cRNA prepared by this method that was used for RNA-DNA hybridization on filters was then treated with Pronase, extracted with phenol, and precipitated with ethanol. If the RNA was not treated with Pronase, it bound nonspecifically to the nitrocellulose filters during the hybridization reaction, presumably because of the small amount of protein associated with it. In Vitro Labeling of cRNA. cRNA was labeled in vitro by the addition of 1 mCi of [3H]dimethylsulfate (New England Nuclear Corp., 100-900 Ci/mol) to 1 mg or less of cRNA in 0.2 ml of sodium phosphate buffer (pH 7.6), made with glassdistilled water (7). Incubation was performed for 6-12 hr at Organ Specificity of Chromosomal RNA 2653 TABLE 1. Protein content of DOC-extracted cRNA from several tissues